heritage reporting corporation

TRANSCRIPT OF PROCEEDINGS

HERITAGE REPORTING CORPORATION Official Reporters

1220 L Street, N.W., Suite 206 Washington, D.C. 20005-4018

(202) 628-4888 [email protected]

IN THE MATTER OF: ) ) ASRAC FANS AND BLOWERS ) WORKING GROUP MEETING ) Pages: 216 through 421 Place: Washington, D.C. Date: May 19, 2015

216

Heritage Reporting Corporation (202) 628-4888

BEFORE THE U.S. DEPARTMENT OF ENERGY OFFICE OF ENERGY EFFICIENCY & RENEWABLE ENERGY IN THE MATTER OF: ) ) ASRAC FANS AND BLOWERS ) WORKING GROUP MEETING ) Room 8E-089 Forrestal Building 1000 Independence Avenue, S.W. Washington, D.C. Tuesday, May 19, 2015 The parties met, pursuant to the notice, at 8:15 a.m. Department of Energy: STEVE FINE ASHLEY ARMSTRONG WADE BOSWELL PETE COCHRAN JOHN CYMBALSKI PARTICIPANTS: KARIM AMRANE AHRI CHRISTOPHER AUTH Baltimore Aircoil Co. MARK BUBLITZ New York Blower Company LARRY BURDICK SPX Cooling Technologies PETER BUSHNELL (Via Webinar) Carrier

217


PARTICIPANTS: (Cont'd) TOM CATANIA AMCA STEVE DIKEMAN AcoustiFLO GARY FERNSTROM California IOUs MARK FLY AAON, Inc. DAN HARTLEIN Twin City Fan Companies ARMIN HAUER ebm-papst Inc. NICHOLAS HOWE Carnes Co. SANAEE IYAMA Lawrence Berkeley National Laboratory DIANE JAKOBS Rheem SAM JASINSKI Navigant Consulting TIM MATHSON Greenheck JOANNA MAUER (Via Webinar) Appliance Standards Awareness Project MICHAEL MCCABE Ingersoll Rand Corp. DONALD MCNEIL Buffalo Air Handling Co. TRINITY PERSFUL Clarage Fan Company ELAINA PRESENT Navigant Consulting

218


PARTICIPANTS: (Cont'd) ANIRUDDH ROY Goodman WILLIAM SMILEY Trane WADE SMITH AMCA, International LOUIS STARR Northwest Energy Efficiency Alliance PHILIP THOMAS Berner Inc. GREG WAGNER Morrison Products MEG WALTNER Natural Resources Defense Council ROBERT WHITWELL Carrier STEPHEN WIGGINS Newcomb & Boyd CHRIS WISEMAN Nidec MICHAEL WOLF Greenheck

219


P R O C E E D I N G S 1

(8:15 a.m.) 2

MR. BOSWELL: Good morning. Welcome to the 3

fourth meeting of the ASRAC Fan Working Group. My 4

name is Wade Boswell and I'm from DOE's Office of 5

Hearings & Appeals and I'm one of the facilitators. 6

For the record, we'll start off as we normally do just 7

going around the room, and if everyone could introduce 8

themselves and identify their organization. 9

MR. BURDICK: Larry Burdick, SPX Cooling 10

Technologies, representing Cooling Tower Institute. 11

MR. THOMAS: Phil Thomas, Berner 12

International. 13

MR. DIKEMAN: Steve Dikeman, AcoustiFLO. 14

MS. JAKOBS: Diane Jakobs, Rheem. 15

MR. WAGNER: Greg Wagner, Morrison Products. 16

MR. HOWE: Nick Howe, Carnes Company. 17

MR. HAUER: Armin Hauer, ebm-papst. 18

MR. BUBLITZ: Mark Bublitz, New York Blower 19

Company. 20

MR. HARTLEIN: Dan Hartlein, TCF. 21

MR. WOLF: Mike Wolf, Greenheck. 22

MR. STARR: Louis Starr, Northwest Energy 23

Efficiency Alliance. 24

MR. FERNSTROM: Gary Fernstrom representing 25

220


the California Investor-Owned Utilities, who are the 1

Pacific Gas & Electric Company, the Southern 2

California Edison Company, the San Diego Gas & 3

Electric Company, and the Southern California Gas 4

Company. 5

MS. WALTNER: Meg Waltner, Natural Resources 6

Defense Council. 7

MR. GOODMAN: My name is Roy Goodman. 8

MR. WHITWELL: Bob Whitwell, Carrier. 9

MR. FLY: Mark Fly, AAON. 10

MR. SMILEY: Bill Smiley, Trane. 11

MR. JASINSKI: Sam Jasinski, Navigant 12

Consulting. 13

MR. CYMBALSKY: John Cymbalsky, DOE. 14

MR. FINE: Steve Fine, Office of Hearings & 15

Appeals. 16

MS. IYAMA: Sanaee Iyama, Lawrence Berkeley 17

National Lab. 18

MR. WIGGINS: Steve Wiggins, Newcomb & Boyd. 19

MR. SMITH: Wade Smith representing AMCA. 20

MR. MCNEIL: Don McNeil, Buffalo Air 21

Handling. 22

MR. MATHSON: Tim Mathson, Greenheck Fan. 23

MR. PERSFUL: Trinity Persful, Clarage. 24

MR. WISEMAN: Chris Wiseman, Nidec U.S. 25

221


Motors. 1

MR. CATANIA: Tom Catania, AMCA. 2

MS. PRESENT: Elaina Present, Navigant 3

Consulting. 4

MR. AMRANE: Karim Amrane, AHRI. 5

MR. MCCABE: Michael McCabe, supporting 6

Trane. 7

MR. AUTH: Chris Auth, Baltimore Aircoil. 8

MR. BOSWELL: Great. Thank you. Unless 9

there are any specific issues people wanted to pick up 10

from yesterday, my understanding is that we're going 11

to start with a presentation by Tim Mathson from AMCA, 12

and I think he's set to go on that. 13

MR. FERNSTROM: So this is Gary for the 14

California IOUs. Before the presentation I have a 15

question that goes back to yesterday. We had that big 16

Trane rooftop unit on the screen and we were talking 17

about the expense and difficulty of testing the air 18

handlers. It literally was about the size of a train 19

car I guess. And I was curious how the manufacturer 20

specifies the performance of the air handler for 21

architects and engineers if it isn't tested. And I 22

was wondering if maybe Wade could speak to how AMCA, 23

you know, may or may not test fans and blowers of that 24

size. 25

222


MR. BOSWELL: And again, if people could 1

identify themselves with their organizations, 2

especially, you know, often -- I thought we did a 3

really good job yesterday of staying more 4

conversational as opposed to, you know, jumping from 5

one topic to the other based on tent cards, but at 6

some point people stopped identifying themselves. 7

And for the sake of the transcript that 8

needs to be made, it would be really helpful if people 9

remember that. And I'll try to remind people, which I 10

failed to do yesterday. 11

MR. SMITH: This is Wade Smith with the Air 12

Movement & Control Association. So AMCA owns and 13

operates several labs, two, one in Singapore, one in 14

the United States. And we do, in the U.S., we do over 15

2,000 tests a year. These are rating tests. So it's 16

not a research lab, it's a performance testing lab. 17

And the question, I'll sort of break it into 18

two chunks to provide an answer. If we were testing 19

the fan, in other words, if the fan assembly were 20

pulled out of the casing and we were to set it up in 21

one of our chambers and test it, typically what's done 22

is that the manufacturer chooses several 23

representative fan diameters and tests them themselves 24

in order to rate an infinite number of operating 25

223


points on many different fan sizes. They don't have 1

to test each fan size. 2

And then if the fan is being certified 3

through AMCA, then if those tests are done in an 4

accredited lab, then there's just a check test that's 5

done, but if not, then basically the entire rating of 6

the product line has to be based on AMCA, tests done 7

in the AMCA lab. 8

So that's how a fan is rated. I should add 9

that we charge our member companies our cost to do 10

these tests. Trane is a member of AMCA. And, you 11

know, a test like that runs around $3,000. Probably 12

my guess is two or three tests would be necessary to 13

rate the entire fan product line, okay? 14

We also do tests of air handler fan sections 15

or air handlers, air handling units essentially, and 16

we do those tests under contract from AHRI as part of 17

their certification program. They have a standard 18

called, I believe it's AHRI 430. Isn't that right? 19

And so, in 430, it's defined what is tested, 20

so I'll just defer to what AHRI 430 says, but 21

nominally, it is the fan section of the unit. It 22

includes some of the components that are inside the 23

air handler but not all. And we test the air 24

performance of that box, which is the fan section, and 25

224


the ratings for air handling units then are published 1

in accordance with AHRI 430. 2

And some of the losses, including the system 3

effects associated with the box around the fan, are 4

implicit in the fan performance tables, and losses 5

which are not implicit and integrated into that 6

performance table are considered external to the fan. 7

In other words, you add the static pressure losses of 8

those components. Am I doing this right, Bill? 9

MR. SMILEY: Yep. 10

MR. SMITH: So far. Okay. 11

MR. SMILEY: Good job. 12

MR. SMITH: Okay. Rooftop units are a 13

little bit different because there's no certification 14

of the air performance of a rooftop unit, so we don't 15

actually test air performance of rooftop units today. 16

If one were to test the air performance of a rooftop 17

unit, you would do something similar to what is done 18

with air handlers. You'd depopulate the unit of its 19

optional devices, right, and then you'd test it 20

without those devices and then the added pressure drop 21

associated with those devices would be included as 22

static pressure losses external to the fan rating, not 23

external to the unit but external to the fan's rating. 24

MR. STARR: So this is Louis. I've got a 25

225


quick question. Louis Starr with NEEA. So by 1

depopulate, you mean things like taking out the -- 2

MR. SMITH: Filters. 3

MR. STARR: -- cooling coils, the heat -- 4

MR. SMITH: Typically not. 5

MR. STARR: Okay. 6

MR. SMITH: But, you know, if they have a 7

two row cooling coil as the minimum, then normally the 8

two row coil would be in the unit, and then if they 9

have a option for a three row coil or a four row coil, 10

the added pressure loss associated with that would be 11

tabled. 12

MR. STARR: Okay. So you try to find the 13

base unit is essentially what you're trying to do. 14

MR. SMITH: Right. 15

MR. STARR: Okay. 16

MR. SMITH: Right. So I think the comment 17

made yesterday about the difficulty of testing the air 18

performance of the unit is valid. However, the air 19

performance of the unit has to be tested in order for 20

Trane to catalog the air performance of their product, 21

and they do catalog the air performance of their 22

product. 23

So the risk to Trane in terms of added 24

testing cost is the need to -- if the regulation is 25

226


written such that there is a dotted line box around 1

the fan, and since Trane, since that fan is not sold 2

as a standalone fan, if the box is drawn around the 3

fan, then the fan has to be -- the performance of the 4

fan has to be determined, and how it's determined is I 5

think a question that's open to discussion, but one 6

way is to remove the fan assembly and its structure 7

and test it in a testing lab. And if we were to do 8

that testing, you know, it would cost a member company 9

around $3,000. We charge more for non-member testing, 10

but for member testing it would be about a $3,000 11

test. 12

MR. FERNSTROM: So this is Gary. Thank you, 13

Wade. The energy efficiency performance of the air 14

handlers constitutes a significant end use use of 15

energy and it's important to us, so I wanted the 16

record to be clear about just what the actual cost and 17

difficulty of determining that parameter is. Thank 18

you. 19

MR. FLY: Wade, this is Mark Fly with AAON. 20

What's the capacity of your chamber or your -- 21

MR. SMITH: I don't know exactly. 22

You know, Steve? 23

MR. DIKEMAN: This is Steve Dikeman. I 24

believe it's 55,000 CFM. 25

227


MR. SMITH: Yeah. 1

MR. DIKEMAN: But that's not anything more 2

than, you know, vague memory. 3

MR. SMITH: It depends on whether the flow 4

measurement is done on the discharge or the inlet of 5

the fan. 6

MALE VOICE: Right. 7

MR. SMITH: But I think it's 55 or greater. 8

MR. FLY: I mean, it depends on the pressure 9

you're testing at obviously. 10

MR. SMITH: Not so much. We can go up to 20 11

inches of pressure differential. 12

MR. DIKEMAN: That's 55,000 CFMs. 13

MR. SMITH: No, because that's a smaller 14

chamber. Right. So I would add, though, however, you 15

know, our members rate and certify the performance of 16

products that are much larger than we can test and 17

they do that using the fan laws, testing a smaller 18

unit and using the results to rate a larger unit. 19

It's a common practice and well-accepted. 20

MR. FLY: And that's kind of where I was 21

going, that a lot of this stuff is projected up from 22

smaller units. And even air handlers a lot of times 23

are rated kind of by an AEDM type process, whereas we 24

have a model that fits a smaller unit well, then we'll 25

228


project it on up to larger units. 1

MR. SMILEY: So this is Bill Smiley, Trane. 2

MR. BOSWELL: Actually, if I could turn to 3

Larry. He's been waiting to make a comment. 4

MR. BURDICK: Yeah. Wade, thanks. I'd like 5

a couple more questions on your test. You said it was 6

$3,000 for a member, one of your members for a test. 7

What is the up charge for a nonmember? 8

MR. SMITH: This is Wade Smith. I think 9

it's double. 10

MR. BURDICK: Okay. 11

MR. SMITH: It's not more than double, so -- 12

MR. BURDICK: Uh-huh. And then the other 13

question is how much more capacity do you have, you 14

know, with your facilities on being able to perform 15

testing? Can you do three times more tests than you 16

do now? Can you do nine times more tests than you do 17

now? 18


MR. BURDICK: How much capacity do you have? 20

MR. SMITH: Right. We run one shift, with 21

an occasional second with a reduced crew. Once in a 22

while we'll have one or two guys on second shift. And 23

so the answer is we could double our testing load with 24

no problem. 25

229


Look, understand, however, that we're in the 1

business of testing and if, you know, the demand 2

exceeded our capacity to supply, we'd make capital 3

investments to increase our capacity. But we're 4

running at about 50, 60 percent of what we could do at 5

the present time. We have, you know, three air test 6

chambers. We can run many tests simultaneously. 7

Dan? 8

MR. HARTLEIN: Sorry. Dan Hartlein, Twin 9

City. I think additional comment to that too is that 10

there is a tremendous testing capacity in the 11

membership of AMCA that is certified lab. I don't 12

know what that count -- I think I saw a number, 50 13

labs. 14

MR. SMITH: Yeah. We accredit laboratories 15

outside of AMCA, and we have 50. Actually, I think 16

the total number is 53. But we have 50 member company 17

laboratories which are accredited to do this kind of 18

testing. We have a lab in Singapore which of course 19

is accredited. We own it. And we have two 20

independent, not manufacturer-owned, two independent 21

testing labs, one in Korea and one in France, that are 22

accredited also. So all of those, all the independent 23

testing labs are available to test people's product as 24

well. 25

230


MR. HARTLEIN: Yeah. Thank you. And I was 1

going to add, Dan, again that most of us have invested 2

in our own labs because it's more cost-effective to do 3

it ourselves as opposed to having AMCA do that, so 4

that would be why we have our own labs. So the number 5

that Wade gave you is somehow kind of a retail number 6

as opposed to an internal cost to a fan manufacturer. 7

MR. SMITH: Right. 8

MR. HARTLEIN: Just to get the numbers 9

right, I would say that test can probably be run for 10

under $1,000 if it's done by a fan manufacturer in an 11

existing lab. 12

MR. SMILEY: This is Bill Smiley, Trane. 13

Just one comment to what Dan just said. I believe 14

that the laboratories that my company has were not 15

built because we can test cheaper than AMCA or 16

somebody else. It's because we do research and 17

development and it's hard to do that on a schedule in 18

an outside lab with all the other pressures on an 19

outside lab. 20

MR. SMITH: Yeah. Very true. 21

MR. SMILEY: So I don't know that we do it 22

cheaper. It might cost more. 23

MR. BOSWELL: Okay. Okay. So what I'm 24

going to ask is if we can put this conversation on 25

231


hold slightly. My impression is that Tim's 1

presentation might be useful for the conversation 2

we're having, so I'd like to move into his 3

conversation and then we can pick up after that, okay? 4

MR. STARR: Well, this is Louis. You know, 5

I think this conversation's actually one of the best 6

ones we've had to date, so I would say we should 7

continue on it just a little bit more. I think what 8

Tim's going to produce, I think most of us already 9

know that on a technical side of things. 10

But one thing, I do have a question I would 11

like to ask Wade that is pretty important for when 12

they drew the box around the fan, and that's can they 13

project -- my understanding is is sometimes you can 14

just test the fan outside of the air handler and 15

project the performance such that you don't 16

necessarily need to actually test the fan inside of 17

the enclosed -- so you're predicting the performance 18

of the fan just with the fan separately, so you reduce 19

your testing burden down even further, but you can 20

project what the performance of the air handler. 21

And that's typically maybe even how they do 22

it. They don't actually bring the whole box in there, 23

but they take the fan and they know how the fan's 24

going to perform in the box. But if you could talk 25

232


about that, that would be pretty helpful I think. 1

MR. SMITH: Right. This is Wade Smith. So 2

the fan laws are a gift from the good Lord himself 3

that gave us the physics, right, that allows us to 4

rate larger products from testing smaller products, 5

but there are limitations. And some of the components 6

which impact the rating of a fan as applied don't 7

scale using fan laws, but that doesn't mean they don't 8

scale. So a manufacturer who rates a product based on 9

the test of a smaller unit has to figure out 10

mathematically, affirmed by testing, you know, how 11

they're going to -- what that relationship is. 12

And such relationships are oftentimes 13

proprietary and unique to a particular product line, 14

and, you know, AMCA is not in the business of 15

affirming, at least we haven't been in the past, 16

affirming people's rating schemes unless they're, you 17

know, in the public domain. Now what we do, however, 18

do is we affirm that their ratings are correct. 19

Even when an accredited lab, like Dan 20

Hartlein was talking about their lab is, and many 21

others have accredited labs, when they bring a product 22

line to market, we still affirm the testing that they 23

did by testing one of the fans that they tested and 24

making sure that the results in our lab are identical 25

233


to the results in their lab so we know that the rating 1

tests that they did at the time that they did it are, 2

you know, affirmed by third-party oversight basically. 3

MR. BURDICK: Larry Burdick with SPX. Wade, 4

it was mentioned here too about certain manufacturers 5

have certified labs. Could you go through that 6

process? What's the cost of the certification, and is 7

there annual licensing, or how does that process work? 8

MR. SMITH: This is Wade Smith. I'm going 9

to correct your semantics just a little bit. When we 10

accredit labs, it's called an accredited lab, okay? 11

So the accreditation process has a cost that, you 12

know, depends on how many trips we make to the lab, 13

depends how well the laboratory is prepared for our 14

visit and whether or not they meet our requirements. 15

It can be as cheap as, you know, 5- or $6,000, it can 16

be as expensive as 20- or $30,000, in that range, to 17

accredit a lab. It depends how many times we go back. 18

And I should add that there are some labs that we've 19

been back to on four or five occasions and they aren't 20

accredited because they just, they don't meet our 21

standards. 22

Certification, right, is a term internal to 23

the AMCA family which refers to ratings, and the 24

certification of ratings is essentially AMCA saying as 25

234


a third-party oversight that we've affirmed that what 1

the cust -- what the customer -- what the member 2

company is publishing for air performance is correct 3

within certain limits. I mean, it's not precisely 4

correct, but the imprecision is defined in a standard, 5

the allowable imprecision. So that's called 6

certification. 7

And if the member has an accredited lab, the 8

certification of a new product line involves one test 9

to affirm the many tests that they did in their lab. 10

They do, however, submit those test results so that we 11

can look at them and assure that they're linked to the 12

tests that we did correctly. 13

If the member company doesn't have an 14

accredited lab, they're going to rate their product 15

line based on tests that we do in our lab. And, you 16

know, how many fans do they choose to test, that's 17

always an open question and it's a question that the 18

member company decides. The more tests that they do, 19

the more fans that they test, the more precise is 20

their rating. 21

However, the imprecision, when you're 22

scaling from a smaller fan to a larger fan, the 23

imprecision is always on the conservative side. So 24

why would you pick more fans? So that your fan 25

235


ratings are better, right? If you pick fewer fans to 1

rate your product line, as you extrapolate to larger 2

units, the resulting ratings that are certified are 3

less than the actual fan will produce. 4

I should just put one caveat in here. The 5

ability to scale these ratings is founded on 6

dimensional similarity, and in this case, I use the 7

word similarity as the technical definition, which is 8

to say that they're dimensionally similar. So that 9

means that the gap between the fan wheel and the 10

housing grows proportional to the size of the fan. 11

Fans aren't exactly dimensionally similar, 12

but they're sufficiently dimensionally similar so that 13

this rating scheme has been proven in many, many, 14

many, many check tests. When we rate a product line 15

or when the member rates a product line, they do it 16

based on testing a few sizes, but then when we do 17

check tests, which we do every three years, okay -- so 18

we pull a fan on certified products every three years 19

to test in our lab and we never pull the same size we 20

tested previously. We try to pull an untested side, 21

size. So, when we pull an untested size, it is in 22

part an affirmation of the fan laws that were used to 23

rate that size. Is that okay? 24

MR. WHITWELL: This is Bob from Carrier. 25

236


Just wanted to add one other thing to the -- you 1

talked about the costs of the tests being 3- to $6,000 2

or something like that, which is in many cases much, 3

many times less cost than the equipment itself would 4

be. So, you know, let's just not think that we've got 5

the test expense. There's also the expense of the 6

equipment under test -- 7

MR. SMITH: Right. 8

MR. WHITWELL: -- which can be substantial. 9

MALE VOICE: And shipping. 10

MR. WHITWELL: Shipping. I mean, we could 11

be talking about approaching $100,000 for some of this 12

stuff. 13

MR. STARR: You could get your lab -- this 14

is Louis Starr. Although like if you're a big enough 15

company you can probably just get your lab certified. 16

You won't need to ship it anywhere, right? 17

MR. WHITWELL: Although I'm wondering, if 18

we're talking about lab certification, I don't know if 19

that's really a requirement, to have a certified lab, 20

right? I mean, we test, do a lot of testing for other 21

energy efficiency metrics, right? 22

MR. STARR: Well, what I meant was is that 23

for AMCA's purposes, they could come and certify your 24

lab, it sounds like, if you had a sufficient thing. 25

237


So you would need to -- if -- so obviously I would say 1

someone the size of Carrier it's probably not going to 2

be a problem for. I would envision that you would 3

certify your lab to AMCA standards. 4

Now it would be a question of what ends up 5

getting negotiated as far as, you know, the 6

ultimate -- 7

MR. WHITWELL: I mean, we've got to see what 8

the test procedure's going to be and everything, 9

right? And again, I haven't heard anything or seen 10

anything in any of the publications talking about 11

anything as far as certification of labs as a 12

requirement, so -- 13

MALE VOICE: It's coming next I think, 14

right? 15

MR. WHITWELL: Thank you. I don't think so. 16

I mean, I don't know. 17

MR. STARR: Well, I was thinking it would be 18

a similar -- well, you know, actually DOE would be the 19

better ones to comment on that. 20

MR. WHITWELL: So I guess, you know, let's 21

see what we -- because you mentioned that you already 22

know about it. So some of us have not seen this 23

material yet, right? So -- 24

MR. STARR: Yeah. Well, I meant this is 25

238


more about systems. It was less about what we're 1

talking about. But it's more about fans and the 2

opportunity to save energy, but yeah. 3

MR. AUTH: I just have one comment. Chris 4

Auth, Baltimore Aircoil. I don't disagree with the 5

geometrically scaling and the rating of fans. This is 6

just a comment regarding the costs associated with 7

testing fans. 8

You know, what we deal with is much larger 9

fans usually and they have special manufacturing 10

processes that aren't typical to, you know, the fans 11

that are tested, that AMCA tests, typically sheet 12

metal or cast type of fans. We're, you know, hand 13

laid fiberglass or extruded aluminum. And to be able 14

to build a fan to be in the say 55,000 CFM limit, that 15

would not be in the product line of our suppliers or 16

even if we made them ourselves. So we would have to 17

go to a modeling shop, make a special fan, and the 18

associated costs to that, I'm not sure, but I would 19

expect it would exceed probably the test cost. 20

MR. FERNSTROM: So this is Gary. Thanks for 21

giving us the opportunity to talk this through. 22

MR. BOSWELL: Okay. So I think we're up to 23

Tim's presentation. And I realize I neglected when I 24

was asking people to identify themselves for the 25

239


record, I think we have at least one working group 1

member that's participating by the web with an open 2

mic, so if I could ask any working group members 3

participating via the web to identify themselves by 4

name and company. 5

MS. MAUER: Hi. This is Joanna Mauer with 6

the Appliance Standards Awareness Project. 7

MR. BOSWELL: Thank you. Any other? 8

(No response.) 9

MR. BOSWELL: Great. Okay. Thank you. 10

Tim? 11

MR. MATHSON: Okay. Tim Mathson from 12

Greenheck Fan. As mentioned, I guess a lot of you 13

have been exposed to this metric that we're proposing, 14

that AMCA's proposing to use for this ruling, but 15

there are some that haven't and especially some of the 16

AHRI folks, and I want to make sure that everybody at 17

least has a view of that and hopefully you can 18

understand how we can use this in your products. 19

This slide is not quantitative at all. It's 20

just what we recognized as the issues that go into 21

saving energy in air systems and just to get them out 22

on the table. And why this is important, kind of the 23

background, we understood going into this that if we 24

simply look at one of these pieces of the pie, like 25

240


fan efficiency, we probably won't have the impact that 1

we want to have, because if we just look at peak 2

efficiency of a fan curve and we increase that, there 3

is a great dependence on this fan selection pie. The 4

size is significant there. 5

Somebody mentioned yesterday that we don't 6

sell fans to building owners who pay electric bills, 7

so there's not a lot of motivation in that fan 8

selection part of the pie. The people who are buying 9

fans, let's say mechanical contractors, are concerned 10

with price and cost, those two things, not efficiency. 11

And so those kind of fight the battle with efficiency. 12

So if we in our effort to increase fan 13

efficiency add cost to the product, we're likely to 14

have people selecting smaller fans yet or less 15

expensive fans yet further away from that peak and 16

ultimately not saving energy. So that's just the 17

background of this. 18

MS. JAKOBS: This is Diane Jakobs. Could I 19

just ask a question about the pie chart? 20

MR. MATHSON: Yes. 21

MS. JAKOBS: You said it wasn't 22

quantitative. Does that mean system design potential 23

is not equal to system effect and system leakage 24

potential? 25

241


MR. MATHSON: We don't have any data to back 1

this up. 2

MS. JAKOBS: Okay. So you just drew it. 3

MR. MATHSON: We know that they are somewhat 4

proportional to their impact. 5

MS. JAKOBS: Okay. 6

MR. SMILEY: Bill Smiley, Trane. I have a 7

follow-up question to that. 8

MR. MATHSON: Yeah. 9

MR. SMILEY: So, if you said there 100 10

percent equals the potential savings, is that 11

representative of what you believe each of those pie 12

sections is worth? 13

MR. MATHSON: Yeah. Yeah. 14

MR. BURDICK: This is Larry Burdick. 15

MR. SMILEY: So, and -- 16

MR. BURDICK: I -- go ahead. 17

MR. SMILEY: No, go ahead. 18

MR. BURDICK: This is Larry and I would 19

disagree with that. You know, one of the points that 20

I brought up yesterday is that system design's much 21

more of interest for the heat exchanger, or what the 22

heat exchanger can contribute to the system dwarfs any 23

of the other sections or items that you have listed 24

there. And I think this slide's real dangerous to 25

242


present to the public, you know, without the caveat 1

that you've identified there, you know, that it's just 2

a representation of maybe the possible effects of a 3

fan system but should not be construed as percentages 4

or capabilities of each particular section. 5

MR. MATHSON: Okay. 6

MR. SMILEY: Yeah. I would suggest just a 7

list rather than a pie chart because the implication 8

of that is that fan efficiency isn't very critical and 9

system, you know -- 10

MR. WHITWELL: I mean, it also says that fan 11

selection is the biggest opportunity and I think that 12

while that may be true in some cases, I think in the 13

case of the HVAC equipment, I don't think that there's 14

that much that can be done across the board to change 15

the fans, the fan types used and improve the 16

efficiency a lot. 17

MR. STARR: So I think what he's referring 18

to is systems that -- 19

MR. BOSWELL: Again, please remember to 20

identify yourself. 21

MR. STARR: Oh. Louis Starr with NEEA. The 22

system he's referring to is the air system, so I don't 23

think he's talking about system in the sense of the 24

air handler, right? Or the -- 25

243


MR. MATHSON: No. Yes. Yes, air system. 1

Yeah. 2

MR. STARR: So this is the ductwork and 3

everything. So it's not -- you're thinking of design 4

as in your air handler and actually what he's talking 5

about is design of the air systems downstream of a fan 6

and not so much really a -- he's not thinking -- this 7

was really not about air handlers. It's more about 8

fan and fan systems that are not -- 9

MS. JAKOBS: So this is Diane Jakobs. So 10

you're only talking about ducted systems? 11

MR. STARR: That's what -- I think that's 12

what you're referring to essentially. And, well, and 13

also, you know -- 14

MR. WHITWELL: But -- 15

MR. STARR: -- warehouses and things like 16

that. 17

MR. MATHSON: Well, for that part, yes. 18

System design? Yeah, that would be ducted systems. 19

Yes. 20

MR. FERNSTROM: This is Gary. So I think 21

the point of confusion here is the fan folks are 22

talking about the air distribution system in 23

buildings, you know, be it what it is, and the heating 24

and refrigeration folks are talking about the fan 25

244


selection and performance within the unit. 1

MR. WHITWELL: Yeah, but I would still say 2

that I still question fan selection being the largest 3

element here. So even applied in the building, I'm 4

not sure that you're going to make much energy savings 5

by changing the types of fans that are used. 6

MR. FERNSTROM: Well, this is Gary again. 7

Our perception is that the energy use associated with 8

ventilation, air movement within the building, is just 9

about equal to the energy use associated with the 10

heating and cooling equipment, so the fan selection as 11

it relates to air distribution within the building is 12

a very important factor with respect to energy use and 13

savings. 14

MR. WHITWELL: No question there's a lot of 15

energy used to move the air around the building. I 16

think -- let's back up. This chart, we should get rid 17

of this chart and list the things that -- 18

MR. MATHSON: Okay. Let me -- okay. Let me 19

just -- 20

MR. WHITWELL: I agree with the list, but I 21

question whether changing the fan selection in doing 22

it is going to -- 23

MR. SMITH: Can I, Tim -- 24

MR. MATHSON: Why don't you go ahead. 25

245


MR. SMITH: -- just one, for a second? This 1

is Wade Smith. Okay. So I'm the guy that made the 2

chart, okay? So this is a perception, the consensus 3

of AMCA members, about where the gold is to be mined 4

for energy efficiency in air systems. The fan 5

selection matters much more than the air dynamic shape 6

of the fan. So just to not be confused by what we're 7

saying here is fan efficiency, meaning changing the 8

aerodynamic shape of the fan, is a small piece of the 9

puzzle. Drive efficiency is another small piece of 10

the puzzle not to be overlooked. It's a significant 11

chunk. 12

Fan selection matters a lot and we can, did 13

actually, when we presented this, demonstrate exactly 14

what the difference is by looking at the fan selection 15

output of any of our members' selection programs which 16

for a given flow and pressure differential will pump 17

out five, six, seven fan selections. And the power 18

consumed by those, to pick a number, five fan 19

selections varies by a factor of two, which is to say 20

that the least efficient fan will use twice as much 21

energy as the most efficient fan. 22

And then we look where does the market 23

actually select these fans. We created a database 24

representing 46 percent of the American market to test 25

246


that question, and they are selected generally very 1

close to the least efficient of those five selections. 2

So the potential for savings by altering selections 3

absolutely overwhelms the potential for saving by 4

changing the aerodynamic shape of the fan. 5

And if you think about aerodynamics, if 6

you're a molecule of air running through the fan, if 7

you go through the fan at half the speed, there's much 8

less turbulence created, and that's the consequence of 9

selecting a larger fan. So it's not wrong. It's not 10

wrong. 11

Now, in the context of an air handling unit 12

or a rooftop unit, when you put a larger fan inside a 13

given casing, what you gain in fan efficiency you 14

might lose in casing losses. But why is the casing 15

size fixed? 16

So again, fan selection matters a lot, 17

system design -- this means the air system design, the 18

size of the components, the pressure drops through the 19

system -- matters a lot, system effects not just in 20

and around the fan, but system effects within the duct 21

system matter a lot. 22

And the point of this chart is not to say 23

that we shouldn't tackle fan efficiency or drive 24

efficiency or fan selection. It's to remind ourselves 25

247


that the goal is not to boost fan efficiency, the goal 1

is to save energy, and there's a lot of energy to be 2

saved from all of these one, two, three, four, five, 3

six areas. 4

System leakage. What percentage of the air 5

sent into the ductwork doesn't arrive at the diffuser? 6

MR. WOLF: So, guys, this is Mike Wolf. If 7

I could for just a second. You know, one thing I 8

think we're all in agreement is this regulation is not 9

going to regulate the building. 10

So, you know, to kind of get back on topic 11

here, Ashley sent me a request last night asking if 12

Tim would, you know, do this presentation this 13

morning. I volunteered him to do so without 14

consulting him ahead of time, so kind of putting him 15

on the spot here. 16

And with that said, you know, the intent, I 17

think, of what Ashley was looking for, probably due to 18

some of my feedback to her, was in participating in 19

the various discussions, you know, the last NODA that 20

just came out introduced a new, relatively new metric 21

to some of us, this FEI. AMCA refers to it as FER. 22

Previously it was referred to as PBER. 23

So anyway, I think what Ashley was hoping to 24

accomplish this morning and have Tim do is just kind 25

248


of give an overview of the FER metric that's in the 1

latest NODA so that we're all at some kind of baseline 2

foundation of understanding what that metric is. 3

Relative to this chart, we could probably 4

debate this forever, you know, and not have any hard 5

data to back it up. You know, it's good discussion, I 6

think. I don't think it's that far off with regard to 7

air systems personally, but I guess I would agree, you 8

know, we probably should just set this pie chart aside 9

for the moment. 10

And, Tim, if you could maybe -- 11


MR WOLF: -- just get into the FER, you 13

know -- 14


MR. WOLF: -- technicalities of the rating, 16

I think, or the metric, that would be good. 17

MR. MATHSON: Okay. And one last comment. 18

I guess when we started looking at an efficiency 19

regulation we started talking about fan efficiency 20

grades, FEGs, which address this one part of the pie, 21

fan efficiency. They don't address the drive 22

efficiency. They don't address fan selection. And 23

what we think we have ended up with is something that 24

addresses each of these three, however big they may 25

249


be, addresses each of the three. 1

MR. HARTLEIN: Tim, Dan Hartlein, Twin City. 2

I also wanted to add that that left half of that chart 3

is more than a thumb in the wind analysis because we 4

did look at a lot of membership data in getting to the 5

assumption of what was happening from a selection 6

practice in the industry overall in looking at the 7

potential savings, looking at that large data set that 8

we were able to compile for a year of shipment. 9

So I think that I would believe that that 10

left half of that chart is probably pretty good. It's 11

pretty representative. Better than, and databased 12

better than perhaps the right half, which was more of 13

where the, you know, the data just doesn't exist in 14

the industry. So we did our best, but -- 15

MR. MATHSON: And it may not be for certain 16

products. Dan, you mentioned -- 17

MR. HARTLEIN: Right. 18

MR. MATHSON: -- process fans where you're 19

building fans to the peak efficiency and they're 20

running at that point. 21

MR. HARTLEIN: We actually have an issue in 22

the larger fans, and we do fans that, you know, will 23

go to 15,000 horsepower in some of our division, one 24

of our divisions. And in the larger fans you have the 25

250


exact opposite problem. What happens here is that 1

conservative design, on top of conservative design, on 2

top of conservative design. Nobody wants a power 3

plant or boiler system to miss its guarantee points 4

because they were short fan. 5

What they end up doing is buying way too 6

much fan to be sure and closing dampers. And we know 7

that's a lot like driving your car with the foot on 8

the brake, right? It doesn't help from an energy 9

perspective. 10

Variable speed and variable frequency drives 11

cures a lot of those ills, so that's a place where we 12

can go in that end of the industry for gains. But, 13

yeah, you're right, on the other end of the business 14

it's the exact opposite problem from a selection 15

perspective. So, but that would be in that selection 16

chart as well. Thank you. 17

MR. WOLF: So, Tim, if I may. Mike Wolf, 18

Greenheck. Just, I can't resist now because Dan 19

opened the door. On that other end of the spectrum, 20

we've done a lot of research in our company as well 21

with regard to fan selection. And as Tim kind of 22

alluded to, you know, the metric that he's going to be 23

sharing with us is really an eloquent way of working 24

fan selection into the metric because what we found is 25

251


that in many cases -- in fact, Tim, you presented, I 1

think you did a forum at an ASHRAE in Dallas, was it, 2

on this subject? 3


MR. WOLF: And you did a scatter diagram of 5

all the fans that we had sold for a year over the 6

various ranges of operation, and the number of times 7

that there was a more efficient fan available for a 8

given selection was a large percentage of the time. 9

And, you know, I don't know if any of you 10

are familiar with the ASHRAE forum process, but it's 11

basically an open mic. I mean, people can come up and 12

say whatever they want to without any, you know, 13

substantiation. So Tim does his presentation and this 14

engineer comes up and he goes, so help me understand 15

why all these engineers are making such stupid 16

selections. And I found it kind of amusing because 17

apparently the guy has never heard of something called 18

bid day. You know, I think I had to just hold myself 19

in the chair to not get up and comment. 20

But, you know, the reason it's happening is 21

the engineer will specify, he'll put a schedule 22

together, specify his fan and then put it out for bid, 23

and every contractor, every manufacturer, you know, 24

they're scrapping to get that job, and largely what 25

252


happens to get that job is you need to be the low 1

bidder, and to be the low bidder you select the 2

smaller fan. And I think Tim will probably have some 3

slides in here that show the difference in size -- 4


MR. WOLF: -- how it affects efficiency, you 6

know, and what we found, if you just would bump the 7

fan up a size or two, you know, the fan will run very 8

efficiently, but that fan's going to cost more, so the 9

guy, you know, at bid day selects as small a fan as he 10

can. 11

And a lot of times it'll probably be outside 12

of the engineer's specification, but he worries about 13

that later, you know, and he'll a lot of times be able 14

to get that smaller fan approved because, you know, 15

the time constraints on the schedule of the building. 16

Well, you know, I've already got, you know, we're 17

three-fourths done, or, you know, with design drawings 18

and I can't go back and change it now, so, okay, good 19

enough. It's such a small element in the overall 20

construction process that, you know, it ends up we end 21

up getting very poor selections and very minimally 22

optimized fan selections on a lot of our projects. 23

So, anyway, sorry, Tim. I'm done now. Go. 24

MR. MATHSON: Okay. Fan efficiency ratios. 25

253


Now FER in this presentation that AMCA has presented 1

is not the same as FER in the NODA that just came out. 2

They switched the terms around a little bit. So just 3

so everybody understands that. This would be closely 4

related to the FEI that's mentioned in the NODA. FEI. 5

And in this whole presentation, I'm going to 6

talk about power in terms of horsepower, shaft power. 7

Eventually this will get into an overall efficiency 8

metric, in other words, wire to air, but it's much 9

easier to talk about in terms of shaft power, so 10

that's what this is going to show. 11

So it's a ratio of fan efficiency to a 12

baseline efficiency, and this is a value that is 13

calculated at every flow and pressure point. The fan 14

efficiency, everybody knows how that's calculated at a 15

certain flow and pressure. The baseline efficiency is 16

also calculated at that same flow and pressure. It's 17

a function of airflow and pressure. So it varies 18

along a fan curve, the FER does, the value of FER. 19

It's independent of the fan type or category. We 20

talked about fan categories here. There are only two 21

test, two different I'll say categories, test 22

configurations, ducted and nonducted. 23

And like, Bill, you said yesterday, we're 24

talking about the discharge. It doesn't matter 25

254


whether it's ducted or not on the inlet side. So 1

whenever we write this we should write it ducted on 2

the outlet, but when we talk about it it's easier just 3

to say ducted and nonducted. 4

And if you think about this baseline 5

efficiency as a minimum allowable efficiency, which it 6

is going to be most of the time -- there may be 7

exceptions to that -- then the value of 1.0 means that 8

you meet the efficiency. Anything greater than 1.0 is 9

you would exceed the baseline efficiency. 10

Okay. So here's a fan curve, pressure 11

curve, power curve. We can plot an efficiency curve. 12

And the whole reason that we're talking about this is 13

the shape of that efficiency curve and how it goes 14

anywhere from zero to a peak value, and where you are 15

on that efficiency curve really determines how much 16

energy is going to be consumed by the fan. 17

So, but with these same variables, pressure, 18

airflow, and power, we can calculate the fan 19

efficiency ratio, and it is a curve, like I said, that 20

varies along the fan curve. Should rise up to some 21

peak and drop off some. And what we want to do is 22

identify the area of that curve that is greater than 23

1, and that becomes the allowable selection range, 24

okay? So we want to define a range on this fan curve 25

255


where manufacturers can actually sell the product or 1

portray the product, its performance. Yes? 2

MALE VOICE: This based on total efficiency? 3

Because you don't have zero efficiency at -- 4

MR. MATHSON: Both. It's based on total 5

efficiency for ducted fans and static efficiency for 6

nonducted fans. So I'm just using it in a generic 7

sense right now. Oh, okay. That's the next slide. 8

So ducted fans -- and this may be a whole 9

separate topic to discuss because this is different 10

from the NODA -- ducted fans can use both the static 11

and velocity pressure to overcome system losses and 12

should be selected using total pressure, and total 13

efficiency then would be the correct measure. Should 14

be selected using total pressure. Most of the time 15

that does not happen, but should be. 16

With nonducted fans, any of the velocity 17

pressure at the fan discharge is dissipated, making it 18

unusable for further work, so nonducted fans must be 19

selected using static pressure, and therefore, static 20

efficiency is the more appropriate measure. Again, 21

that could be a whole separate discussion. 22

So the baseline efficiency, the bottom of 23

that ratio, so a baseline or minimum efficiency varies 24

with airflow and pressure. These are lines of 25

256


constant efficiency, increasing as you go up in 1

airflow and up in pressure. So this is one way to 2

look at it. 3

At reduced or at -- airflow is proportional 4

to fan size essentially, so larger systems are to the 5

right. And pressure can be thought about as 6

increasing fan complexity. You know, to develop 7

higher pressures you need to add things to fans, like 8

scrolls or straightening vanes or different things. 9

Those increase the efficiency at higher pressures, but 10

they decrease the efficiency at lower pressures. So 11

this pressure component here is really to get rid of 12

fan categories so that we can use the same metric to 13

talk about prop panel fans that we do for centrifugal 14

blowers. 15

Another way to look at this same thing is a 16

3D plot here, airflow and pressure. And this baseline 17

efficiency again takes on some shape here. And 18

anything above this surface would be FER greater than 19

1, anything below it would be less than 1. 20

MS. JAKOBS: Excuse me. This is Diane 21

Jakobs from Rheem. And just, it's probably a dumb 22

question, but where does this relationship come from? 23

So you're drawing curves. Is it something tested? 24

Is it fan laws? 25

257


MR. MATHSON: Yeah. This baseline -- so 1

this baseline is a function of airflow and pressure. 2

The airflow part of it -- 3

MS. JAKOBS: No, I can see it's a function 4

of it. 5


MS. JAKOBS: But where do you get it? 7

Where did -- 8

MR. MATHSON: Where did it come from? 9

MALE VOICE: It's a derivation. 10

MR. MATHSON: Well, the airflow relationship 11

is exactly the same as the original FEG curves, and 12

that reflects how fan efficiency drops off with 13

smaller fans. 14

MS. JAKOBS: So did you test 1,000 fans and 15

this is the result or? 16

MR. MATHSON: Yeah. Essentially, yes. 17

Yeah. It's based on what's available on the market. 18

MS. JAKOBS: And how it performed on the 19

AMCA 210 test? 20

MR. MATHSON: Yes. Yeah. 21

MS. JAKOBS: So just a straight duct or a 22

straight -- the airflow is straight, right? 23

MR. MATHSON: No. Any different type of 24

fan. There were many fans that we plotted the peak 25

258


efficiency versus the fan size or the airflow and that 1

took on that shape. 2

MR. STARR: So this is Louis with NEEA. The 3

basis is the brake horsepower equation is what it is. 4

And essentially they've added some scalers by looking 5

at their fan selection. So they take the brake 6

horsepower equation and put a scaler in for the flow 7

and put a scaler in for the static pressure, and then 8

the ratio of a baseline fan is one to the other. So 9

it's really just using the brake horsepower with some 10

scalers added in is the basis of the FER. 11

MR. SMILEY: Bill Smiley, Trane. Just a 12

quick question, Tim. So you had a significant amount 13

of data that you used to develop the numerical values 14

of all this stuff and that significant amount of data 15

was based on using every fan type you had data on, 16

which would have been airfoil, BI, BC, radial, axial. 17


MR. SMILEY: And that's probably about it 19

for AMCA. So it doesn't really have FC fans in there, 20

it doesn't really have panel -- 21

MR. MATHSON: FC fans, yeah. 22

MR. SMILEY: It does? 23


MALE VOICE: Everything. 25

259


MR. SMILEY: So, so, and then the result of 1

that would be a huge, wide curve of data. 2

MR. MATHSON: Right. Right. Right. 3

MR. SMILEY: And what you did there is you 4

took this wide curve of data and you said here's the 5

curve we're going to assign to all fans and make it 6

the same for everything. 7

MR. MATHSON: That's right. That's right. 8

MR. SMILEY: Is that basically what you did? 9


MR. SMILEY: Do you have that scatter chart 11

that shows here's the range, or do you get into that a 12

little bit later? 13

MR. MATHSON: No. 14

Go ahead, Wade. I'm having a hard time. 15

It's not as simple as that. 16

MR. SMITH: Right. So it all started with 17

FEG. When the committee was working on what the FEG 18

formula should be, there was a recognition that 19

smaller fans are less efficient than larger fans, 20

right? And so the fan committee, the fan engineering 21

committee did scatter plots of different fan types to 22

determine what that relationship was and then 23

reflected it in the family of curves which define the 24

fan efficiency grade. 25

260


And what you do is you take the fan 1

efficiency and diameter and you go into that chart and 2

then you figure out what the FEG is. And the shape of 3

those banana curves, so to speak, is reflective of 4

those scatter diagrams. Those scatter diagrams are 5

all fan types on the same diagram with a curve fit, 6

okay? 7

MR. SMILEY: So a least squares fit? 8

MR. SMITH: Well, yeah, essentially. So 9

there's two -- this was FEG and the FEG curves were 10

much debated for years, yeah, and adopted. So, when 11

Tim crafted this approach, he used those curves to 12

deal with flow. And the shape of this curve, if you 13

were to make a slice through it -- he's got it on the 14

screen right now -- and just look at baseline 15

efficiency versus flow, you'd see it's the FEG curve. 16

He did the same thing with pressure, okay, 17

and the idea of creating a diminished requirement for 18

smaller pressures, that's how the constant is added 19

into the formula. The idea of creating a reduction in 20

the required efficiency for fans operating at lower 21

pressures was an accommodation for the lower 22

efficiency that's implicit in low pressure operation. 23

And the types of fans optimized to minimize the power 24

consumption at low pressures also diminishes the peak 25

261


efficiency of that fan. 1

So there's a pressure constant of 0.4 in 2

this equation and there's a flow constant of 250. The 3

flow constant mimics the FEG curves. The pressure 4

constant is a recognition and an accommodation to fans 5

designed specifically for operating at low pressures 6

to reduce their efficiency requirement. And there's a 7

tradeoff. As you reduce the efficiency requirement 8

for fans operating at low pressure, in order to 9

generate the savings from any particular regulatory 10

level, you've got to boost the efficiency requirement 11

at the higher pressures, right? And so the membership 12

debated at length what accommodation should exist, 13

okay? And so they did this with the benefit of a lot 14

of data in front of them. 15

But to say it's a curved fit is not correct. 16

It's the consequence of a debate which intended to 17

diminish the efficiency requirement of low pressure 18

applications vis-à-vis high. The other one is, after 19

years of debate about FEGs, is simply an extraction of 20

that relationship. So now we have a three-dimensional 21

curve, right, that has both of these things at work. 22

MR. AUTH: Chris Auth, Baltimore Aircoil. I 23

have two comments. 24

One is, it goes back to yesterday with the 25

262


whole discussion with the embedded fans. So this 1

whole database as I understand it is an AMCA database. 2

So I don't have a clear understanding of that 3

database, how many embedded fans would be represented 4

in it. I know our products, the percentage is 5

probably close to zero. I would suspect it's maybe 6

higher than that for smaller fans, but I would suspect 7

it's a pretty low number. You know, I'm not doubting, 8

you know, the fan types we use are similar. I'm just 9

saying that database doesn't represent a good sample 10

of embedded fan types that are used today in the 11

market. 12

And the second point I have is it goes back 13

to -- you know, it's the thought that sometimes people 14

would select a less efficient fan to save money, to 15

lower costs. That's not always the case. There's 16

some -- you know, engineering, you know, it's not just 17

about efficiency. There's also like one -- one big 18

example would be sound. We use fans that are less 19

efficient that, you know, cost us several times more 20

than the most efficient fan, but we use it because of 21

sound for our customers. They have a sound limit that 22

they have to have. So that's just one point I want to 23

bring up. 24

MR. MATHSON: Steve. 25

263


MR. DIKEMAN: Steve Dikeman. I'm going to 1

argue with Tim, but I'm saying it with a smile on my 2

face. This graph right here illustrates a particular 3

set of inputs illustrating where FER is greater than 1 4

or less than 1. 5

MR. MATHSON: That's right. 6

MR. DIKEMAN: Right? And so now, as we look 7

at different fan categories, classes, the target 8

efficiency might be different, probably will be 9

different. So, if you were to go draw another graph 10

with a different target efficiency, it would move up 11

and down in the vertical axis where you meet 1.0, 12

exceed 1.0, or fall below. So this isn't -- it was 13

compared against our database, but this is just one 14

individual set of inputs. Your target efficiency in 15

this graph is a single number, correct? 16


MR. DIKEMAN: And so, if you had a different 18

target efficiency for a different fan category, that 19

graph would change. 20

MR. MATHSON: That's correct. 21

MR. DIKEMAN: That's the simplicity and the 22

subtlety of FER. If this fan category has this target 23

efficiency and if you took the baseline efficiency off 24

your graph, now it would apply to the conversation 25

264


we're having. You've made an individual input to 1

create this graph. You know what I'm saying? 2


MR. DIKEMAN: So, for a different fan class 4

with a different target efficiency, it wouldn't go up 5

to 63 percent baseline. It would be higher or lower. 6

Target efficiency changes with fan class, category, 7

whatever the correct moniker is for that. 8


MR. DIKEMAN: 1.0 and higher is a compliant 10

product. 11

MR. MATHSON: I want to say something else 12

to maybe help lighten the -- so that maybe you're not 13

quite as afraid of this as you are. 14

If you are personally looking at fan 15

selections, you're probably exceeding this 16

significantly, okay? If you are a mechanical 17

contractor, you may not be exceeding this, okay? And 18

Wade talked about the database. We looked at all the 19

different fan types that were in the database and if 20

we put this surface here let's say, we get let's say 21

20 percent of the fan selections were below this 22

surface for this fan type and for this fan type and 23

for this fan type. We tried to match up so that, you 24

know, the same percentage would be below this curve 25

265


that would be above, and like I said, it's a small 1

number and if you're paying any attention to the fan 2

selection, you're probably well above this surface. 3

MR. JASINSKI: So, Tim, I'm just going to 4

interrupt for a second. I think this is a good 5

conversation and it's important to understand what 6

assumptions and ultimately what data went into the 7

approach that's being presented, but I think maybe 8

what we should try to do is get a little bit further 9

into this approach because I'm worried that even 10

though they're related that we might be debating FEG, 11

and that's not what we are here to do. We want to 12

talk about the merits of the approach later on. 13

I think the message being sent here is that 14

there is a relationship with efficiency and airflow 15

and pressure. You can't just set one efficiency 16

target at all airflows and all pressures because the 17

way fans work, there are -- you know, it's 18

inherently -- fans are inherently less efficient 19

across a range of airflows and a range of pressures. 20

So, if everybody is in general agreement 21

about that, not the quantitative aspect but maybe just 22

the qualitative aspect, I think that's going to play 23

into how the metric is structured, and then once we 24

get to the actual metric being discussed then we can 25

266


say okay, well, for the mechanisms in the metric, in 1

the FER or the FEI metric being discussed, let's look 2

at the data that goes into that and we can tweak those 3

factors or those aspects of that metric as opposed to 4

all the stuff leading up to it. 5

So let's get to the end and then work our 6

way back maybe a little bit. I think that would be a 7

better way to approach this. 8

MR. MATHSON: Wade, you have something to 9

add. 10

MR. SMITH: This is Wade Smith. I just want 11

to respond to the question about embedded fans. No, 12

there are many embedded fans in the database. And the 13

other point that you asked about I think also deserves 14

mention, and that is that we did not focus on nor seek 15

to include forward-curved fans that were in regulated 16

unitary equipment because at the time that we pulled 17

the database together we were of the belief that they 18

weren't, you know, part of the rulemaking. So there 19

are a lot of embedded fans missing, right, because we 20

didn't look at that group, but we did go back to our 21

membership and affirm that there are lots of fans that 22

they sell OEM which become embedded that are included 23

in the database. 24

With respect to cooling tower axial fans, 25

267


the answer is that we've got lots of axial fans in 1

here, lots of prop fans, lots of panel fans, but I 2

don't think too many of them truthfully went into 3

cooling towers. So I think to the extent that we 4

acknowledge, I think it's important to acknowledge 5

that we don't have a lot of cooling tower fans in here 6

and we don't have a lot of fans that are embedded in 7

regulated equipment in the database, but we do have 8

lots of embedded fans. 9

MR. MATHSON: Okay. So the FER can be, as I 10

said before, the ratio of fan efficiency to baseline 11

efficiency. It can also be flipped over and talked 12

about in terms of power because they're the same 13

variables here. So the same fan efficiency ratio is a 14

baseline power or in this case it would be a maximum 15

power, allowable power, divided by fan input power. 16

That's a little bit easier to work with. 17

So here's the form of the equations. Like 18

Sam said, this is just to get us going here to 19

understand them. For fans tested with a ducted 20

outlet, the total efficiency, the baseline total 21

efficiency is some target value, and Steve mentioned 22

this determines the height of that surface curve. 23

There's an airflow factor and a pressure factor, and 24

the Q-naught and the P-naught are constants that are 25

268


in these equations. And then for nonducted the whole 1

thing is done in static pressure and static 2

efficiency, same form. 3

So that's the form of the equation. Again, 4

the values in blue are constants, and these are some 5

of the constants that we think are real good. The 6

target efficiencies, again, have yet to be 7

established, and -- 8

MR. DIKEMAN: That's what this meeting is 9

about. 10

MR. MATHSON: Yup. 11

MS. JAKOBS: So this is Diane Jakobs from 12

Rheem again. So, just to be clear, 250 and .4 are 13

constants selected by the AMCA committee members? 14

Okay. 15

MR. MATHSON: Yes. So this is just a 16

little -- 17

MR. WOLF: Tim, this is Mike. I've sort of 18

waited. It wasn't just AMCA, right? We did this with 19

the energy advocates. Were they part of that process 20

as well or no? 21

MR. SMITH: This is Wade. Yeah. I mean, 22

the big debate was amongst AMCA members to decide what 23

to recommend, and then we had to explain why and what 24

the impact was and why the numbers should be higher or 25

269


shouldn't be lower. So I'm not sure. Did the 1

consensus comment that we made to the NODA, did it 2

include these constants? I believe it did. 3

MS. MAUER: This is Joanna. I don't think 4

it did. I think we said that -- we had open 5

discussions about it, but not an agreement. 6

MR. SMITH: Open to question, yeah. 7

MR. FERNSTROM: So this is Gary. The 8

efficiency advocates were part of the discussion where 9

this concept was presented, and to the extent that it 10

addresses a little broader system approach toward 11

application we were in support of it, and it follows 12

the same line of thinking that the Hydraulics 13

Institute utilized for pumps based on the European 14

pump standards work. 15

MR. ROY: Aniruddh Roy, Goodman. I have a 16

question for either Tim or Wade. If you can just go 17

back to the previous slide on the constant. Just 18

following up on Diane's question. So since the sample 19

size did not include forward-curved impellers, would 20

the constant change if they were included in the data 21

set to determine the equation? 22

MR. SMITH: This is Wade. It did include 23

forward-curved. It did not -- we made no effort to 24

try and include forward-curved fans that are embedded 25

270


inside regulated unitary equipment. So your question 1

then would be if those were included, yeah, would the 2

numbers change. I really doubt it. 3

MR. MATHSON: I don't think so. 4

MR. SMITH: But, you know, if we ever get 5

the data, we'd be happy to analyze it. 6

MR. MATHSON: And this graph just shows 7

where each of those constants affects/impacts the 8

shape of this. The target efficiency on top again 9

raises or lowers this whole surface. The curvature on 10

the sides are impacted by the Q-naught and the P-11

naught. 12

So, if this is what the metric looks like, 13

what would the regulation look like? So we've got a 14

metric here, FER, and the ways, different ways that we 15

could use this. The federal regulation could require 16

fans to be sold greater than 1.0 at both its peak and 17

at its design point. You know, peak is kind of 18

obvious there, but at its design point if that 19

information is give to the manufacturer. 20

ASHRAE 90.1 could require FER greater than 21

1.0 at the design points. Same concept. 189.1 could 22

require a greater number, a 10 percent increase in 23

efficiency over 90.1. 24

And then rebates, if we have utility 25

271


rebates, we can use this as a basis for those rebates. 1

So we've established a baseline at 1.0. If you've 2

got a fan for a particular operation at 1.2, you know 3

you're going to use 20 percent less power, and then 4

you go through the calculations on what that rebate 5

might be for that purpose. 6

So this is how it can be used, you know, in 7

a regulation sense, but to a person who is selecting a 8

fan, I think it's much more powerful to see this 9

value. 10


Rheem again. So your design point, it would be 12

defined as a pressure? 13

MR. MATHSON: Pressure and an airflow. 14

MS. JAKOBS: And an airflow. 15


MALE VOICE: It needs efficiency of power 17

too, doesn't it? 18

MR. MATHSON: Well, the customer would come 19

to the manufacturer and say I need an airflow and a 20

pressure. So, yeah, to calculate the FER, you need 21

power and the efficiencies. 22

MR. WHITWELL: So one of the things that we 23

have to talk about in that case is how do we define 24

the pressure and the airflow on a product where that 25

272


gets determined by the application, right? So our 1

customers tell us what they -- they determine what the 2

pressure requirements are for the product, the 3

external static pressure requirement. They tell us 4

what the airflow is. 5

MR. DIKEMAN: For every curve. 6

MR. WHITWELL: Right. 7

MR. DIKEMAN: It's the duty of the FER. 8

MR. WHITWELL: But this is in a product 9

that's already designed and available for sale, and 10

they can vary it, right? 11

MR. BOSWELL: So, again, please remember to 12

identify yourself. 13

MR. WHITWELL: Bob Whitwell from Carrier. 14

Thank you. Sorry. 15

So, I mean, we just need to understand how 16

this would be applied in a situation like that, right? 17


MR. SMITH: This is Wade. You just 19

described every sale of every fan, right? Yeah. 20

MR. SMILEY: Bill Smiley, Trane. 21

MR. MATHSON: Yeah. Go ahead. 22

MR. SMILEY: But with an OEM type product 23

the customer or user defines the external static 24

requirement. He doesn't know or probably doesn't even 25

273


care what the total pressure requirement for the fan 1

is. He's looking at his system needs to match up to a 2

unit that outputs pressure to overcome his system 3

requirement. 4

MR. SMITH: That's true, Bill, of every -- 5

MR. SMILEY: So, I mean, that's where we get 6

into a sticky spot here. 7

MR. SMITH: That's true, Bill, of every fan. 8

A fan sold standalone, the statement you just made 9

would apply. 10

MR. SMILEY: Mm-hmm. 11

MR. SMITH: So the question is at the 12

customer's flow and pressure what is the maximum 13

horsepower allowed, and that's the question that we 14

hope to answer in the regulation. So, for the given 15

flow and given pressure, what is the maximum 16

horsepower that the DOE will authorize? 17

MR. MATHSON: Let me answer that in a little 18

different way. If a custom air handler manufacturer 19

goes to a fan manufacturer and orders a fan at a 20

certain airflow and pressure, it includes both the 21

external pressures and the internal losses in the air 22

handler, right? So the customer is the air handler 23

manufacturer. 24

In the case of a packaged unit, the customer 25

274


is specifying external static pressure, and so what 1

you're saying is, yes, somehow we have to correlate. 2

They don't know what the internal pressure drop is, so 3

we have to work on that. So that's in the 4

administration of this I guess. 5

MR. JASINSKI: Tim, I just want to make sure 6

if what you're describing is accurate. I mean, what 7

I'm hearing is that even though the end user of maybe 8

a rooftop unit who's specifying the external static 9

pressure and the flow, that the manufacturer of that 10

RTU can take that and translate it into what does one 11

of my fans internally need to produce in terms of flow 12

and pressure, because the RTU manufacturer will know 13

what those internal losses are and they'll be able to 14

translate that and give that to the fan manufacturer 15

who is supplying the fan in general. 16

MR. MATHSON: Well, they don't have to run 17

that test. They don't have to. They could run the 18

test of the air handler with the fan in it. 19

MR. JASINSKI: Right. 20

MR. MATHSON: And just measure the external 21

performance. 22

MR. HAUER: It's Armin Hauer speaking. I 23

would assume that the fan static pressure cannot be 24

known inside of the unit because inlet affects and 25

275


outlet system affects and you cannot distinguish it to 1

an embedded equipment. So what you really know for 2

that fan in that application is only the flow. The 3

static pressure, fan static pressure you cannot know. 4

MR. MATHSON: Okay. These are the right 5

questions and they get to how would this look in an 6

OEM product. 7

MR. WHITWELL: So just one more comment. 8

Bob Whitwell from Carrier. So you talked about having 9

to understand the end customer's requirements and then 10

selecting the fan based upon that. So what really 11

happens is we might have a couple different fans 12

selected and qualified in the product. So it's not a 13

situation where a customer tells us they have this 14

sort of requirement there, out there as far as static 15

and flow, so we go look and see, okay, where does it 16

fall on the third number, and say, okay, we go to a 17

fan manufacturer, we need a new fan to put in this 18

unit because they're running at a condition where it 19

drops below the one. 20

So, in that case then, we would have to go 21

back in and requalify that fan in our equipment. If 22

it's got gas heat, we've got to put that fan in. 23

We've got to make sure that the airflows are all still 24

good and the safeties all work, and this is months of 25

276


stuff that would have to be done. So it does create 1

some difficulties that we have to work around and 2

understand. Either that or we're going to have to say 3

we can't sell you that unit. We can't sell you that 4

unit. I don't know if that's what we want to do. 5

MR. SMITH: This is Wade Smith. No, that's 6

not the situation that we're describing here. The air 7

performance of a unit under this scheme would have a 8

compliant range of operation, and I think if you look 9

at the data and look at the curves and look at the 10

proposed efficiency levels, I think the outcome of 11

that analysis is that what you just described would 12

not happen, all right? 13

So I just think you have to be patient and 14

get through the analysis and do the comparison with 15

your own product line to appreciate that, but that's 16

the analysis that our members went through, and what 17

we did was we compared different regulatory schemes, 18

different curve shapes, and different regulatory 19

levels against the database of actual sales, and then 20

begged the question how many selections were 21

noncompliant relative to any level or shape that you 22

might want to ask about, and how would the customer 23

cure the noncompliance? How would the rep basically 24

cure it? And then the follow-on question is, is that 25

277


an undue burden? And the other follow-on question is, 1

how much energy do we save? 2

And that analysis has all been done for many 3

of the products, and the recommendation that we came 4

up with is a reasonable burden to fan manufacturers 5

and produces a lot of savings. So it works. Does it 6

work for your product? You need to analyze it. 7

MR. WHITWELL: Okay. So -- 8

MR. SMITH: What you described is something 9

that doesn't work. 10

MR. WHITWELL: Yes. 11

MR. SMITH: And if it didn't work, it 12

wouldn't be acceptable. 13

MR. WHITWELL: Yeah. So I was just 14

responding to the comment that was made that -- my 15

understanding, maybe I misunderstood what he said, but 16

it sounds like that could happen. So I'll be patient 17

now and listen to the rest. 18

MR. MATHSON: Okay. So this metric will 19

serve not just to regulate fans but also to teach or 20

to encourage good fan selection. So what we envision 21

is that the FER value will be shown in catalogs where 22

performance is shown. So, in this case, the green 23

highlighted area is all greater than FER 1.0, and 24

whatever we do with that, obviously we would not be 25

278


able to sell fans down in that white area that's not 1

highlighted, but it also, you know, tells you how far 2

you are away from the peak. 3

If you'll notice, the numbers are lower here 4

as you get up into higher pressures. They're up in 5

the 1.2s at six inches, but at one inch they're up as 6

high as greater than 2, and you'll see that a little 7

bit more with the fan curves why that is. The 8

efficiency requirement, the baseline efficiency is 9

lower at lower pressures, so the FER becomes -- well, 10

I guess they're not over 2. They're about 1.4, 1.5. 11

But in electronic fan selection software, a 12

customer will put in a design point of operation, in 13

this case 10,000 CFM at three inches of pressure, and 14

they will get a list. In this case, this was 15

centrifugal fans with eight different selections, and 16

again, like Wade mentioned, the smallest fan is 12 17

horsepower, the largest fan is six horsepower. The 18

lowest number is the 33, size 33. So we're showing 19

the power and the baseline power, so, again, baseline 20

power would be maximum power allowed. That would 21

equate to an FER of 1, and then the FER value on the 22

right-hand side. 23

MR. SMILEY: Tim, Bill Smiley, Trane. 24


279


MR. SMILEY: A quick question. The largest 1

fan, 36, you're saying is an acceptable selection is 2

implied here, but the peak efficiency, your actual 3

total efficiency dropped off a little bit. 4


MR. SMILEY: That's to the left of the peak 6

efficiency point? 7

MR. MATHSON: It's to the left of the peak, 8

yes, so it's right next to the -- 9

MR. SMILEY: But it's not in stall. 10

MR. MATHSON: Right next to it. 11

MR. SMILEY: Not in stall yet. 12

MR. MATHSON: Yup, not in stall yet. 13

MR. SMILEY: Okay. 14

MR. MATHSON: Right on the edge. 15

MR. SMILEY: But damn close. 16

MR. MATHSON: Not a good selection. 17


Rheem. So, if you had a space-constrained application 19

and you needed a 20-inch fan size, there's just no 20

solution? 21

MR. MATHSON: Well, there is, there is. You 22

need to get a more efficient fan. 23

MR. SMILEY: There's not one -- 24

MR. MATHSON: Different model, different 25

280


model. 1

MR. BURDICK: How do you propose to raise 2

the efficiency from a .8 FER to over 1? That's 20 3

percent efficiency. 4

MR. MATHSON: Well, if this wasn't scrolled 5

centrifugal to start with, I would say go to a more 6

efficient fan. But this is an efficient fan to start 7

with. You won't get there, to answer your question. 8

MR. FLY: This is Mark Fly at AAON. One of 9

the things that I'm a little concerned at, maybe 10

you're going to get to it, but I have yet to see a 11

system curve plotted on any of these fan curves. You 12

know, the trend and what we've been driving to in many 13

of our regulations is to try to go to variable speed 14

fans. In a multi-zone VAV system, the system curve 15

does not start at zero-zero. It starts at some 16

operating pressure that it takes to run the boxes. 17

MR. MATHSON: Right. 18

MR. FLY: So, as you turn the fan speed 19

down, you go into stall at some point. As you pick 20

the fan selection at design where it runs 1 percent of 21

the time or less, at the very peak of the curve, which 22

this is all driving you up to, then you have less 23

turndown in that fan. 24


281


MR. FLY: So I'm really concerned that we're 1

going to encourage -- I mean, we don't always select 2

smaller fans because they're cheaper. Sometimes we 3

select them to have more turndown in a VAV 4

application. 5

MR. MATHSON: Yeah. Yeah. The way I would 6

recommend personally addressing that is to have a 7

lower FER requirement for VAV systems. 8

MR. JASINSKI: Yeah, Tim, I'm just going to 9

-- so I think a lot of -- 10

MR. MATHSON: But we haven't gotten into 11

that yet. 12

MR. JASINSKI: Right. I think a lot of the 13

questions we're hearing are going to be good questions 14

and things that need to be addressed and considered 15

once we start selecting levels in that we don't -- 16

what I'm hearing from the working group is that we 17

don't want to select levels that may completely 18

eliminate certain fan types, certain operating points, 19

and trigger a lot of these costs that, Bob, you 20

brought up, correct? 21

But I don't think that we can -- to Wade's 22

point -- I don't think that the working group can 23

really assess or evaluate those things until we are 24

talking about an actual proposed level, which we're 25

282


not doing here right now. Right now we are simply 1

talking about a metric, not a specific -- you know, 2

not a baseline total efficiency. You know, for 3

instance, for this particular product maybe the 4

correct baseline efficiency is below 59.4 and all of 5

these would still be viable selections, but that's not 6

commenting on the appropriateness of the metric 7

itself. 8

So I think for now, you know, maybe that's 9

because everybody is on board and in agreement that 10

this is a good metric and we're ready to start talking 11

about what appropriate levels should be, I don't know, 12

but I think for the working group's sake and for the 13

sake of doing future analysis what we're trying to 14

accomplish today is establish a baseline understanding 15

of the metric that was proposed by AMCA, a very 16

similar metric that was used in the NODA by the 17

Department of Energy, and get comment on whether or 18

not we think that's a workable solution, regardless of 19

where the levels might end up. 20

MR. WHITWELL: Yeah, Bob Whitwell, Carrier. 21

So good comments. I think, though, that the 22

questions still -- we need to get the questions out 23

there, right, and it's part of the discussion to 24

understand what some of the implications of this 25

283


metric could be, right? 1

MR. JASINSKI: Absolutely. 2

MR. WHITWELL: So I think we need to ask 3

them now, understanding that there will be more -- 4

we'll have to talk about them some more when we get to 5

that level. 6

MR. JASINSKI: No, absolutely useful. I'm 7

not suggesting that we shouldn't have had that 8

conversation. I'm just saying in addition to those 9

questions we also do want to get some feedback on the 10

approach itself, understanding that there are those 11

concerns and, you know, it's part of understanding 12

this to understand the impacts. Point well taken. 13

MR. MATHSON: Yeah, I would say if your -- 14

that question, that particular question from Mark, 15

that means you understand what we're talking about 16

here. 17

MR. SMITH: Mark, there's a lot of debate 18

inside AMCA members about how to deal with VAV, and 19

there needs to be an accommodation, so it's a 20

recognized need that you described, and so, you know, 21

at some point we should put it on the table and talk 22

about it. 23

MR. FLY: This is Mark Fly with AAON. I 24

just bring it up because I have seen hundreds of fans, 25

284


maybe thousands of fans, running in stall in my career 1

from people trying to do the right thing and fans in 2

stall -- I mean, FC fans, you can run in stall all 3

day. You start running BI fans in stall and things 4

start flying apart. 5

So I think it's very important that we not 6

only just look at the design point selection, but, you 7

know, if you look at the IEER weightings, which is 8

basically looking at building loads, it's been well 9

vetted over time, that's telling you that these 10

systems are running between 50 and 75 percent the vast 11

majority of hours that they're operating in a VAV 12

system, so we have to make sure that it works across 13

that range. 14

MR. WHITWELL: Right, and it's not just the 15

VAV, but, I mean, in the other working group where 16

we're looking at commercial unitary air conditioning 17

efficiency, I mean, the IEER metric is driving us to 18

fan staging, so varying fan speeds, even without 19

having it go all the way down to a full VAV type 20

situation. So we can't put a metric here on a fan 21

that will not allow us to get to the overall system 22

efficiency levels, or else we're just going to shut 23

the whole industry down. 24

MR. FLY: Right. 25

285


MR. MATHSON: Okay. Let's look at some more 1

pictures. Here's a multi-speed fan curve with a very 2

efficient fan. There will be a large portion of that 3

map that is greater than FER 1.0, and this is not 4

actual data. This is just an artist's rendering here. 5

This is what it will look like. The more efficient 6

the fan is the larger percentage of this performance 7

map will be highlighted or allowable to sell. 8

But what I wanted you to notice about this 9

is remember the baseline efficiency increases with 10

increasing airflow and pressure. So, at low airflows 11

and pressures, in other words, low speeds, you can use 12

the whole fan curve. The whole curve is green. The 13

whole curve is allowable. As you get up into higher 14

pressures and flows, the baseline is increasing and so 15

the allowable range of operation has shrunk. 16

MR. SMILEY: Tim, Bill Smiley, Trane. I 17

have a quick question on this. So what this is 18

telling me is that the base efficiency varies along 19

the system curve. 20


MR. SMILEY: Okay. And that's why the FER 22

varies. 23


MR. SMILEY: And does not follow a system 25

286


curve. 1

MR. MATHSON: The fan efficiency would be 2

constant, but the baseline is changing. 3

MR. SMILEY: Okay. 4


MR. SMILEY: See, that's why I was confused 6

yesterday. It would have been a simple response. 7

Then I read your stuff last night and I go, oh, well, 8

hell, that's how it is. Thanks. 9

MR. MATHSON: Let's talk about an 10

inefficient fan. Okay, it may have a range of 11

operation that is much smaller. Again, this is just 12

an artist's rendering here. I just changed the 13

numbers. Today this product might be catalogued 14

beyond its allowable range, okay? And so a change -- 15

a regulation like this may restrict this fan from 16

being sold at these higher powers or higher pressures 17

and flows. 18

MS. JAKOBS: This is Diane Jakobs. Another 19

dumb question. So, when you say it restricts the 20

sale, so that would mean the manufacturer's literature 21

would say its application was only within a certain 22

range? 23

MR. MATHSON: Mm-hmm. 24

MS. JAKOBS: And then there would be no 25

287


enforcement. People buy stuff and do whatever they 1

want, right? I mean, I don't -- 2

MR. MATHSON: We can't control where exactly 3

the fan is going to operate in the field. We can 4

control where it's designed to operate, where it's 5

sold or intended to operate. So you can talk about 6

labeling and things, Sam. 7

MR. JASINSKI: Yeah, I'll jump in. I think 8

naturally where this conversation is going to go are 9

compliance enforcement labeling questions. Ashley is 10

on her way and I think she should be in the room for 11

that discussion. So I think -- 12

(Laughter.) 13

MR. JASINSKI: Oh, there she is. Well, I 14

was going to suggest maybe -- yeah, I was going to 15

suggest maybe we take a break if everybody is ready to 16

take a break, and be prepared after maybe a 15-minute 17

break to comment on whether or not this is the metric 18

we should be using going forward in the analysis, and 19

if the answer is no, why not and what should we be 20

using. 21

MR. HAUER: Can I just tag on a question 22

about Slide 17, the relatively efficient fan? Right. 23

Okay. Now let's assume your black curve, the very 24

far to the right, this is a direct drive fan with, I 25

288


don't know, four pole motor, right? But now your 1

customer needs a performance point between the curve 2

to the right and then the next one down. Right, right 3

smack in between. How do you accomplish that? 4

MR. JASINSKI: Between here? 5

MR. HAUER: And what type of an FER would 6

you declare there? 7

MS. JAKOBS: Between 1 and .9? 8

MR. MATHSON: Right where the arrow is? 9

MR. HAUER: Yeah, between -- yeah, exactly 10

there. How would you arrive there, you know, if you 11

have a direct drive fan? 12

MR. MATHSON: With a VFD. 13

MR. HAUER: Okay. So where is the -- 14

MR. MATHSON: Is that what you're asking? 15

Is that your question? 16

MR. HAUER: Okay. So where is then the 17

efficiency impact of the VFD or maybe the belt 18

reflected? There are some losses associated with 19

varying the speed to that point. 20

MR. SMILEY: Well, you could pick the next 21

size smaller down if that was the correct increment. 22

MR. HAUER: Yeah, but there is a likelihood 23

that you will never exactly measure design points. So 24

basically I'm leading to the question where do you 25

289


reflect the losses in the belts and in the variable 1

speed drive and the motor impact from a variable speed 2

drive? 3

MR. WOLF: This is Mike Wolf with Greenheck. 4

So I think we're going to get into somewhere down the 5

line here, Tim, the discussion of how we get from wire 6

to air? 7


MR. WOLF: Right? 9


MR. WOLF: I guess, Sam, I'll just ask him. 11

How many slides do you have left? 12

MR. MATHSON: Maybe 10; not important ones. 13

MR. WOLF: Do you get into the -- 14

(Laughter.) 15

MR. MATHSON: They're all pictures. 16

MR. WOLF: Do you get into the 207? 17

MR. MATHSON: No. 18

MR. WOLF: You don't? 19

MR. JASINSKI: We have some slides that -- 20

MR. WOLF: Okay. 21

MR. JASINSKI: -- reflect the wire-to-air 22

aspects which I think will address your questions. 23

MR. BOSWELL: I was going to say there is a 24

proposal to take like a 15-minute break. We've been 25

290


going for almost two hours. So unless there's an 1

objection why don't we do that. 2

(Whereupon, a short recess was taken.) 3

MR. BOSWELL: Okay. If we could all come 4

back to our seats. So we're going to pick up with the 5

balance of Tim's presentation and then continue the 6

discussion. 7

MR. MATHSON: Okay. So here's -- we talked 8

about a -- I'm going back up. This is what an 9

efficient fan might look like in its performance map, 10

what an inefficient fan might look like in its 11

performance map. We talked about -- we've talked 12

about at AMCA what we do when we don't know the actual 13

design point. So, if a fan is sold through a 14

distributor or something like that, the regulation may 15

say that the FER must be greater than 1.0 at the best 16

efficiency point at the maximum RPM. So, in this 17

case, this fan does not meet that requirement, but it 18

would meet that requirement if the max RPM was just 19

reduced to a certain level. So I kind of think this 20

could be part of the regulation as well. 21

MR. SMILEY: Tim, Bill Smiley, Trane. Quick 22

question. When you say "max RPM," you mean max design 23

capability like the fan, the next fan up? 24

MR. MATHSON: Well -- 25

291


MR. SMILEY: No. It says FER must be 1 or 1

greater at best efficiency point at maximum RPM. 2


MR. SMILEY: You mean the fan capability RPM 4

or what? 5

MR. MATHSON: Well, that would be part of 6

the regulation. I would say it could be the range 7

that the fan is able to be advertised at. 8

MR. SMILEY: Okay. So you would say your 9

max allowed usage RPM. 10


MR. SMILEY: Not necessarily design limit or 12

anything like that? 13

MR. MATHSON: Right, right. 14

MR. SMILEY: Okay. Thank you. 15

MR. MATHSON: So, you know, in this little 16

sort of made up example here, today let's say we 17

published beyond that speed. If there was a 18

regulation tomorrow and we could only go up to a lower 19

speed, that would probably take away some of the motor 20

horsepowers for that fan, so we might redesign it 21

smaller, you know, a little bit differently, or we may 22

take it off the market, you know, and just replace it 23

with something else. 24

And if this regulation was the only 25

292


regulation, then we would just change the maximum RPM 1

to meet that. If there is something else, like let's 2

say the regulation said for safety fans you can use an 3

FER of 0.9 or if we wanted to use this fan overseas 4

somewhere, you know, those are all really fine details 5

about how it could be administered. 6

But one example here as we end up, this is 7

just to illustrate what we want to accomplish with 8

this, with this metric. Here's a square in-line fan 9

that we sell a lot of that is not very efficient. 10

It's just a centrifugal fan inside of a box, and it's 11

ducted inlet, ducted outlet. Sometimes you see them a 12

lot as a return fan or an exhaust fan. It's 13

inexpensive, easy to install, easy to maintain. It's 14

all about the contractor. It's made for the 15

contractor to install, but it's low efficiency. 16

If we look at the performance map for this 17

fan, this is a 30-inch wheel in a square in-line fan, 18

and now because it's ducted I'm showing total pressure 19

here. This is the range that we catalogue today, and 20

so it's got a best efficiency point, a maximum total 21

pressure 53 percent, which is really up high on the 22

fan curves. It's kind of unique that way. But you 23

can draw other lines of efficiency. You can see how 24

it dropped, the total efficiency drops as you go down 25

293


the fan curve. 1

And we've plotted actual fan selections from 2

the year 2012 on this map to see where people are 3

selecting these fans. And so, as you look at this and 4

where the efficiencies are, I think it becomes obvious 5

that these customers don't care very much about 6

efficiency. It's all over the place. There's no 7

grouping here anywhere near high efficiency. There's 8

a lot of selections down towards the end. 9

And if, if we were going to have this 10

regulation of FER greater than 1, that allowable 11

selection range would be as I have shown here. So 12

about 40 percent of these fan selections for this 13

model fell within this, and this is just what we had 14

proposed as some base efficiency levels, target 15

efficiency levels. So about 40 percent of the 16

selections met the requirement. 17

So the purpose of this is, what do you do 18

with the other 60 percent? Well, if I look at one 19

size larger, size 36, that covers more of that range, 20

so that covers 70 percent now of the fan selection. 21

So 40 percent of them could have been done with the 22

30, the next 30 percent I guess of selections could be 23

covered by one size larger, and then about 90 percent 24

of them could have been covered by the next size 25

294


larger. So some of the fans had to go one size up, 1

some of the fans had to go two sizes up. There are 2

some of the fans that are higher pressure there that 3

are not covered by any of those bubbles. Those would 4

have to be covered by a different fan type. This fan 5

doesn't cover that. 6

But if you look at this. So I looked at one 7

point out here and said well, what's the customer 8

going to do; 15,000 CFM at a half-inch of total 9

pressure. So his option here is to use a larger fan 10

size, a size 42, but that's pretty big. So here's 11

that design point, 15,000 CFM at .5 inches. The size 12

30 was -- here's the details on that: 5.3 brake 13

horsepower; FER .62, so that was not an acceptable 14

selection there, the size 30, and some of the other 15

details, operating cost, budget cost. 16

So, because of that, they can replace that 17

with a size 42, so 42-inch wheel. Quite a bit larger 18

in size to get that FER above 1.0. So it moved up to 19

1.12, and you see the FER is proportional to the 20

power. The FER went up by almost a factor of 2. The 21

power was almost cut in half. But it's a heavy fan, 22

735 pounds. It's more expensive. You could, and I 23

put on there the housing widths, so the original was 24

46 inches wide, this was 48 inches wide. So the 25

295


customer may say, well, that's too large, I can't fit 1

it into the space that I want. 2

So you can replace that with a mixed flow 3

fan, a size 27 will do this, with an FER greater than 4

1. So its horsepower again is 2.77, about half of 5

what the original selection was. But if you look at 6

the cost, it's twice the cost of that original 7

selection, and that's why we didn't sell this fan in 8

this application. 9

So this is the customer's choice, but what 10

we're trying to do here is to get the customer to come 11

back to the fan manufacturer and say, hey, I need a 12

new square in-line fan with better efficiency, okay? 13

I can't accept this 58-inch square. I need something 14

smaller, so redesign this. We want our customers to 15

say redesign this fan to get a better efficiency or 16

come out with a new mixed flow fan that's less 17

expensive that competes with that. And either one 18

would be a good answer for that situation. 19

So, again, the point of this illustration is 20

we don't want to -- you know, if we're just -- if the 21

regulation is just concerned with efficiency and not 22

the fan selection, we'll design a more efficient 23

square in-line fan, but they'll still pick the size 24

30, so we're trying to capture all of that together 25

296


here. 1

What is this? Oh, this is something else. 2

Extended products, I won't talk about this. 3

MS. ARMSTRONG: Why don't we move -- 4

MR. MATHSON: Yeah, you can talk about that. 5

Okay. That's all I have. 6

MR. SMITH: This is Wade Smith. While 7

Sanaee's team get up I thought I would just add a 8

comment. It was mentioned that when we started 9

working on this question of how to measure fan 10

efficiency we started down the path of peak 11

efficiency, our FEG, which was an accommodation to 12

smaller fans that have lower peak efficiency, and the 13

membership of AMCA got quite far down the road with 14

FEG, wrote standards, recommended FEG nomenclature for 15

90.1, et cetera. 16

And then we switched, and the reason we 17

switched is because we had the assignment to calculate 18

the savings associated with different FEG levels, 19

pushing assimilation basically or a query into the 20

database and asking for, hey, what is the savings, 21

what is the impact, how many fans are going to have to 22

be redesigned, et cetera. 23

And then a debate ensued about the method of 24

calculating the savings, and out of that discussion 25

297


grew a realization that our calculation of savings was 1

highly speculative and that the savings that we did 2

calculate was at risk over customer behavior, 3

responding to a higher priced fan, like what was just 4

shown, by instead purchasing a smaller size. 5

And so the question of what savings would we 6

generate from a peak-based efficiency metric became -- 7

you know, turned into a big question mark, and the 8

common expression was that the change in peak 9

efficiency was not deterministically linked to a 10

change in energy use. This is a statement which is 11

more true for some fan categories than others. It 12

depends on how broadly across the spectrum fans are 13

selected, and we looked very carefully at scatter 14

diagrams where fan selections actually occurred, and 15

another way to look at is this: that if the fan 16

selection varies from 5 percent to 85 percent and you 17

raise the peak efficiency by two or three percentage 18

points, you know, what have you really done? 19

In other words, you've got two things 20

working to drive energy use. One has tremendous 21

leverage and the other one has very little leverage. 22

The one that has tremendous leverage needs attention. 23

And so that's when we went back to the drawing board 24

and Tim Mathson and the folks at Greenheck made this 25

298


recommendation and suggested that there was a 1

deterministic link between the efficiency rating at 2

the design point and the actual energy consumed, and 3

we went on from there. 4

So the suggestion and the proposal that grew 5

out of that was to have DOE regulate the efficiency of 6

the fan at its design point, not at peak efficiency, 7

and that's kind of where we are and why, in terms of 8

the recommendation that we made, why we ended up 9

there, just so everybody understands. 10

MS. ARMSTRONG: Yeah, so that's a pretty 11

good segue into what Sanaee is about to present to 12

you. Before we start, Peter online wants to comment 13

about the last example. Before we move on we might as 14

well. Unmute him. 15

Peter, you should be unmuted. 16

MR. BUSHNELL: Yeah, hi. 17

MS. ARMSTRONG: Hi. 18

MR. BUSHNELL: Thanks. It's Peter Bushnell 19

at Carrier. Just on the last example, I found that 20

one interesting because as we look at operating point 21

15,000 CFM, 0.5 inches of water gauge duty pressure, 22

if we look at all of the fan classes and types for 23

that duty point and kind of look at a specific speed-24

based approach across all fan types, you'll find that 25

299


the optimal choice of fan is the axial fan, and I 1

found it interesting that that didn't come out of the 2

analysis that was done. So was I wondering if you 3

could comment on that a little bit, Tim. 4

MR. MATHSON: This is Tim Mathson. Yes, I 5

would agree that it looks like more of a tubeaxial 6

application, but why wasn't a tubeaxial on that slide, 7

I can't answer that question. 8

MR. WAGNER: This is Greg. I would say that 9

was a selection thing. 10

MR. BUSHNELL: The reason I bring that up is 11

that -- 12

MR. WAGNER: Sorry. This is Greg Wagner. 13

MR. BUSHNELL: The reason I brought that up 14

was that if you follow simply kind of looking through 15

catalogs and databases you'll come up with one 16

solution, but if you use kind of a systematic approach 17

using specific speed approach, which I think was what 18

was used in the DOE pump regulation for efficiency, 19

then there's kind of a more systematic way to find 20

what's the right type of fan for a given duty point, 21

and this is the most critical point in choosing a 22

given fan. 23

And, Ashley, I'm having a lot of feedback, 24

by the way, so this is very challenging for me to 25

300


actually speak here. 1

MR. HARTLEIN: So, Peter, this is Dan 2

Hartlein from Twin City Fan. We would agree. There's 3

actually theoretically a range where the specific 4

speed would indicate a typical axial selection, that 5

as specific speed changes things like hub ratio and 6

blade solidity change in the design of the product. 7

There is obviously as well a specific speed where a 8

centrifugal fan tends to rule the day. And then 9

there's products where the customers in the market 10

want to buy something that is different than both 11

because it's cheap and easy and it works for them in 12

the field, and I think that that centrifugal fan in 13

the box is an outcome of that. 14

That is absolutely correct that that's 15

operating at a specific speed which would typically 16

and technically be an axial fan, and it's a duty point 17

that should be there, but the market is buying a 18

product or demanding a product that is not -- I think 19

Tim rightfully showed a peak efficiency of that 20

product line at 53 percent. That is not a product 21

that's being designed and sold for its energy 22

efficiency. That's clear, and I think you're 100 23

percent right that that would be a, from a specific 24

speed question, that would be a axial fan. You're 25

301


absolutely right. 1

MR. BUSHNELL: And I think that the example 2

also brings up another interesting point, is that 3

there may be for that example other constraints like 4

stall margin performance or extended range of 5

pressurized; you know, going down, down to lower flow 6

rates, and with those kinds of constraints, you know, 7

it might be that, hey, the mixed flow fan is really 8

required, but if you're strictly looking at .5 at 9

15,000 CFM, you know, without those extended 10

requirements, axial fan is probably the right choice. 11

MR. HARTLEIN: Yeah, I think we agree. I 12

think you had your hand up. 13

MR. BUSHNELL: So this discussion on design 14

point, I think those are the kind of factors that come 15

into play on design point. You know, it's what are 16

the extended range requirements, stall margin, things 17

like that. 18

MR. WIGGINS: This is Steve Wiggins from 19

Newcomb Boyd. I think the other thing we need to 20

realize too when we start talking about the customer 21

or the -- and I'm going to step across the line and 22

say the designer who is specifying the product, 23

specific speed means nothing to him in his world 24

currently, okay? And so, when you start looking at 25

302


how do we get to the point where we're making the fan 1

selection, I think we have to realize too that if we 2

go certain directions we're talking about potentially 3

training an industry to do their business in a 4

different manner than what they're currently doing, 5

and that may be a bag of worms we don't want to go 6

play with too hard. 7

MR. BUSHNELL: Yeah, it's a challenging area 8

because in a perfect situation you would have tools 9

and selection processes and a regulation that somehow 10

is consistent with that specific speed methodology 11

because that would drive the optimal selection, and I 12

think the problem is the industry doesn't really have 13

that completely formalized for fans. We don't have 14

universal specific speed charts that are used 15

throughout the industry, so it makes it difficult. 16

MR. HARTLEIN: Yeah. Steve, just one other 17

comment to that. That world that you describe on 18

specific speed is a driver for selection actually 19

happens in the heavy-duty industry, so our Clarage 20

division will select a fan and design a fan optimized 21

to the specific speed, the duty point, but that's a 22

custom fan. We start to see that happen in higher 23

horsepowers. Some markets will demand that fan at 150 24

horsepower, so we do get into the 200 horsepower range 25

303


that we seem to be talking about here. But it's 1

driven by the fact that you can have a million dollars 2

of energy consumption a year on a 10,000 horsepower 3

fan, and in that industry, it does in fact function in 4

the way that you're describing, but the HVAC industry 5

clearly does not at this point. 6

MS. ARMSTRONG: Okay. So I'm going to make 7

a suggestion at this point that we're going to walk 8

through our metric. It is more or less the AMCA? 9

Advocates' metric that they represented in the NODA. 10

At least we are going to explain it and what this is 11

translated to look like in kind of the DOE world and 12

how this would work. We've highlighted some points 13

for discussion that we would like some feedback. 14

We're hoping, though, that at least some of the stuff 15

we can think about voting on this afternoon in terms 16

of high-level metric and some of the test procedure 17

type issues. So, Sanaee, why don't you start walking 18

us through some of the metric slides and we can have 19

some discussion. 20

MS. IYAMA: So I'm going to go by the first 21

slides because I think Tim went over this already. 22

Fan efficiency varies greatly depending on the fan's 23

operating point. Here it's just illustrating the fan 24

efficiency distribution amongst different fan 25

304


selections from the fan data we got from AMCA. And 1

so, to address this issue, DOE used a metric evaluated 2

at each operating point as declared by the 3

manufacturer, so it's an approach similar to the one 4

presented in AMCA's white paper, and it's based here 5

on the wire-to-air approach, so that's the difference 6

with what Tim presented earlier, and it's based on 7

determining the electrical input power of the fan at a 8

given operating point, what we called FER in the NODA. 9

So the FER would include the fan shaft input 10

power, which is the first part of that equation, flow 11

times pressure divided by fan efficiency. If there's 12

a transmission, like a belt, that would be also 13

incorporated in the electrical input power to the fan, 14

as well as driver losses and control losses if 15

controls are used. 16

I'm just going to go through the slides, or 17

do you want to stop? 18

MS. ARMSTRONG: So, yeah, hang on. So 19

before you go through -- so what do we think about 20

this? 21

MR. SMILEY: Bill Smiley, Trane. The FER 22

you're referring to really is a horsepower or watts? 23

It's not a ratio of any kind. It's a value. 24

MS. IYAMA: It's just electrical input 25

305


power, so you could express it in watts or horsepower. 1

MR. SMILEY: Well, the term FER is a little 2

misleading then. I mean, you ought to just call it 3

watts. 4

MS. ARMSTRONG: So we've teed that up. 5

Probably unknownst to the fan industry, DOE does have 6

an FER metric already for residential furnace fans. 7

So what do you propose it be called? 8

MR. SMILEY: Well, call it watts. That's 9

what it is. 10

MS. ARMSTRONG: How about fan electrical 11

power? 12

MR. SMILEY: I don't care. 13

MS. ARMSTRONG: FEP? Can you deal with 14

that? Great. I mean, it's either that or Ashley. I 15

don't know. 16

MR. SMILEY: I think I'll vote for Ashley. 17

MR. FINE: While you're at it, you could add 18

a C to the transmission units. 19

MR. SMILEY: We don't care about spelling. 20

MS. ARMSTRONG: Yeah, we don't. That's 21

really not -- 22

MR. SMILEY: We're engineers. 23

MS. ARMSTRONG: Exactly. That's not 24

important to engineers. The lawyers will take care of 25

306


that during the months of review. 1

MALE VOICE: I don't know if I should have 2

caught that. 3

MALE VOICE: These aren't going to turn 4

into -- these aren't going to be words in any way. 5

MS. ARMSTRONG: Yeah, exactly. 6

MR. HAUER: It's Armin Hauer speaking. I 7

was terribly confused about the horsepower, that we 8

called it -- it's electrical power, but we call it 9

horsepower. And horsepower I think is like mostly in 10

the United States. It's a mechanical power. Okay. 11

Can we please -- 12

MALE VOICE: It's brake horsepower. 13

MR. HAUER: I've never seen electrical power 14

expressed in the unit horsepower. 15

MS. IYAMA: So I think in the NODA it 16

doesn't really come up because we ended up doing the 17

ratio. So it has no dimension. 18

MR. HAUER: But in the spreadsheet -- 19

MALE VOICE: Where do you see horsepower 20

here? 21

MR. HAUER: In the spreadsheet, it's 22

calibrated back to horsepower. 23

MALE VOICE: So you're suggesting working in 24

watts when using electricity? 25

307


MS. ARMSTRONG: So would you prefer using 1

watts always? Okay. No. Well, you guys figure it 2

out in that corner. 3

MR. HAUER: But when it's electrical power, 4

it's watts, right? 5

MALE VOICE: Yeah. 6

MS. WALTNER: This is Meg. If we changed 7

the term to fan input power, would that -- 8

MALE VOICE: It's not fan power. 9

MALE VOICE: Let's just change the noun. 10

MS. ARMSTRONG: Right. I would suggest 11

changing the acronym in that case, but -- 12

MR. SMILEY: Bill Smiley, Trane. It's not 13

really fan input power. It's power into the upstream 14

device that connects to the fan through a drive and a 15

motor and on and on and on and on. So the fan power 16

is always in the industry referred to as the actual 17

brake horsepower or the power that's consumed by the 18

rotating device. 19

MS. ARMSTRONG: So why is it that you want 20

to use horsepower? 21

MR. SMILEY: Let's be confused here. 22

MR. HARTLEIN: Because it's -- I think 23

everybody -- excuse me. Dan Hartlein. I think 24

everybody that's in here that doesn't have European 25

308


origins is comfortable with horsepower as that 1

expression. It's no offense, but that's a metric 2

issue. We use it there. We all know what it means. 3

MR. BUBLITZ: Well, Mark Bublitz here. 4

MR. HARTLEIN: Technically he's right. 5

MR. BUBLITZ: But when we talk horsepower, 6

we talk at the shaft -- 7

MS. ARMSTRONG: Yeah. 8

MR. BUBLITZ: -- so that you can size the 9

motor. 10

MR. HARTLEIN: Not all of us. We don't. 11

MR. FERNSTROM: So this is Gary. I think 12

the Europeans have this figured out, but we haven't 13

quite figured it out yet. And generally horsepower 14

here is deemed to be mechanical, and electrical would 15

be watts or KW, and that's what I would prefer. And 16

we're talking about the electrical input power to the 17

motor controller system that drives the mechanical. 18

MALE VOICE: Yeah. 19

MR. BURDICK: This is Larry with SPX. I 20

have a comment about each operating point. There are 21

so many operating points associated with a fan. You 22

know, is it at the design point for that particular 23

cooling, you know, for the design cooling duty? You 24

know, then there are all the stream of points that are 25

309


associated with when you're running on VFD, you know, 1

or off-duty situations, different motor loadings, you 2

know, different heat rates and so on. You know, I 3

think a broader term or a consistent term might be 4

peak fan efficiency. 5

MS. ARMSTRONG: It's not peak. 6

MR. BURDICK: No? 7

MS. IYAMA: So here it would be at each 8

operating point as declared by the manufacturer. 9

MS. ARMSTRONG: So it wouldn't just be peak. 10

It would be all. 11

MR. BURDICK: So this is an operating point 12

declared by the manufacturer as to which point it is 13

I'm talking about. 14

MR. SMITH: It's the point of sale. 15

Somebody comes to -- this is Wade Smith. When 16

somebody comes to buy a fan, they say, I need a fan 17

that will do the following duty. So we're selling the 18

fan for operation at that duty at a certain efficiency 19

level. Now where does the customer operate the fan? 20

MS. ARMSTRONG: Kind of. Kind of that. 21

MR. JASINSKI: Wade, I think, I think the 22

answer is -- 23

MR. SMITH: Kind of that? Okay, kind of 24

that. 25

310


MR. JASINSKI: -- everybody is right. 1

MR. SMITH: Okay. 2

MR. JASINSKI: What we're talking about here 3

is that it's a -- everybody here is right, but what 4

we're talking about here is the metric allows for 5

these things to be calculated at any operating point. 6

This is not a declaration of which ones we're talking 7

about. We're just saying we are using a metric that 8

regardless of who determines it, whether it's the 9

consumer, the manufacturer, the regulation, that you 10

can determine compliance with this metric regardless 11

of what that operating point is. There are no 12

operating points at which you operate a fan that you 13

can't calculate the FEP as it's written right now. 14

MR. BURDICK: So the each should be an any? 15

Any I mean. 16

MR. JASINSKI: Any. 17

MR. BURDICK: Rather than each? 18

MS. ARMSTRONG: Correct. 19

MR. JASINSKI: At any operating point is 20

probably better. 21

MR. BURDICK: Okay, yeah. 22

MR. JASINSKI: And then we will get into the 23

discussion about what those operating points are and 24

how they're determined later. What we're just trying 25

311


to do is make sure that we have a metric that allows 1

for FEP to be calculated at any operating point. 2

MALE VOICE: So any specific operating 3

point? 4

MS. ARMSTRONG: Any is fine. 5

MR. FLY: Yeah. This is Mark Fly. This is 6

not the point of regulation. This is just something 7

to get there. 8

MS. ARMSTRONG: This is just a metric. 9

MR. FLY: Just a metric to get there. 10

MS. ARMSTRONG: Do we agree with the 11

components of the metric? 12

MALE VOICE: Well, let's see. 13

MS. ARMSTRONG: So, I mean, just generally 14

speaking, this is similar to the approach you guys 15

have worked on. We'll change the name, FEP. 16

MALE VOICE: To AER? 17

MS. ARMSTRONG: You can come up with a 18

better name next week or in two weeks or in August. 19

That's fine. But generally these are the main 20

components that are included in an overall metric. 21

MR. WHITWELL: So the motor efficiency is 22

included in the transmission? 23

MALE VOICE: No. Driver losses. 24

MS. ARMSTRONG: Driver. 25

312


MR. WHITWELL: It's in the driver losses? 1

So that's driver and motor losses I guess. Is that 2

yes? 3

MS. ARMSTRONG: Yes. Okay. Moving on. I 4

think we have a thumbs up. 5

MR. DIKEMAN: Ashley? Ashley, Steve 6

Dikeman. Before you go, your discussion point on 7

bearings, they're included in AMCA fan ratings if that 8

fan has a fan shaft through it. So you've already got 9

them. 10

MS. ARMSTRONG: But if they don't, they are 11

not. 12

MR. DIKEMAN: It's a direct drive. You have 13

no bearing losses. 14

MS. ARMSTRONG: Right, okay. 15

MALE VOICE: It's part of the motor 16

efficiency. 17

MS. ARMSTRONG: Right. 18

MS. IYAMA: So here fan electrical input 19

power, so for any operating point it could be 20

determined based either on measuring the fan shaft 21

input power and then combining that with default 22

values to represent the motor, the transmission or 23

controls, if any. So that's the first option. We 24

called it the calculation-based method. Or it could 25

313


be directly measured at the input of the electric 1

motor controls, if any. So that would be the testing-2

based method. 3

MR. SMITH: This is Wade Smith. Would you 4

then also allow a test that includes the transmission 5

but not the motor, for example, and use default values 6

for the motor and controls? In other words, is there 7

a continuum between those two extremes, or is it just 8

the two extremes? 9

MS. ARMSTRONG: I mean, I think we could. 10

Do we need to? 11

MALE VOICE: Do it anyway. 12

MR. SMITH: I don't know. I just wanted to 13

know where your stand was, and then I guess we can ask 14

our members what they want. 15

MS. ARMSTRONG: Yeah. I think it's more a 16

question of what do you want. 17

MR. SMITH: Okay. 18

MS. ARMSTRONG: I don't think we have a -- 19

MR. SMITH: So you would have no problem if 20

we step through that as an option. 21

MS. ARMSTRONG: Not necessarily, no. 22

MR. SMITH: Okay, good. 23

MS. ARMSTRONG: I mean, such that they 24

provide equitable ratings. 25

314


MR. SMITH: No, no, I understand. 1

MS. ARMSTRONG: But no. 2

MS. IYAMA: Okay. So then this metric would 3

be compared to what Tim has presented this morning, 4

but he presented it in terms of maximum allowable 5

shaft input power. Here we're talking about maximum 6

fan electrical input power. So same thing. It would 7

be based on an equation of efficiency as a function of 8

flow and pressure, and then we would combine that with 9

default values for motor, transmission. And here, for 10

that determination, we wouldn't consider controls. 11

One thing that was in the AMCA white paper 12

was the issue of how to evaluate fans with controls. 13

I think AMCA had presented in that white paper the 14

possible inclusion of sort of a multiplier to the FER 15

standard here as it's called when evaluating fans with 16

controls. 17

MR. SMILEY: This is Bill Smiley, Trane. A 18

comment there. If, for example, you -- oh, I pushed 19

twice. Sorry. Bill Smiley, Trane. For example, if 20

you had a motor with a VFD on it, variable speed 21

control device controlling the speed of the motor, the 22

motor efficiency changes depending on the output and 23

the type of controller you're using to control the 24

motor. So how would you accommodate or account for 25

315


that inefficiency? Would that be on the motor, and 1

then it varies at every load point on the motor and 2

setting point from the controller? Or would you 3

account for that in the controls? How would you 4

account for that? 5

MS. IYAMA: For the losses of the VFD? 6

MR. SMILEY: There are two things here, 7

losses through the VFD and the effect on the motor of 8

the VFD signal going to the motor that changes the 9

motor's efficiency relationship with speed and 10

frequency and the wave form, the type of motor and the 11

type of VFD, and the list is very long of those 12

variables. 13

MS. IYAMA: So here we're in that scenario 14

where we're trying to establish the input power of the 15

fan, and it has controls. And so the first option is 16

you could do a wire-to-air test. That's the second 17

bullet on that slide. Or the first option is you 18

establish sort of a default model that allows you to 19

calculate the losses of your motor and VFD, the 20

combined motor and VFD. AMCA has a draft 207 where 21

they have one recommendation on how to do that. We're 22

working also trying to develop a similar or just 23

evaluating their approach, and I think that's probably 24

something that we'll have to discuss. 25

316


MR. FLY: This is Mark Fly at AAON. I 1

realize that, you know, what you're talking about 2

changes -- is changeable by punching the buttons on 3

the keypad of the VFD. If you change the switching 4

frequency, if you change the load profile curve on 5

that, it changes the motor efficiency. And that's 6

something that happens in the field all the time. 7

MS. IYAMA: Okay. So before going into sort 8

of the more detailed stuff, just one slide here about 9

using an index. So, in the NODA, we kind of followed 10

what was in the AMCA white paper and took, you know, 11

the baseline electric -- or the maximum allowable 12

electrical input power, divided that by the electrical 13

input power of the actual fan, and called that the 14

FEI. And that's exactly what Tim has presented this 15

morning, except that here it's a wire-to-air-based 16

index. 17

Now there might be issues with using an 18

index. One of them could be the number of significant 19

digits. Maybe two significant digits may not be 20

enough to differentiate between higher variation in 21

fan efficiency. So, if you're trying to compare fan A 22

and fan B, you know, they might have the same FEI if 23

you use two significant digits, but they may not have 24

the same electrical input power usage. So that's one 25

317


example of the issues that could come up with using an 1

index. 2

From a standard level-setting perspective, 3

if the maximum allowable electrical input power 4

changes, so let's say you change that target to a new 5

higher target, then, you know, that index doesn't 6

really mean anything anymore. You could have fans on 7

the market that are calculated with a different 8

baseline because the regulation has been updated. 9

MS. JAKOBS: This is Diane Jakobs from 10

Rheem. So I don't know if really the issue is 11

multiple digits. It's more like multiple operating 12

points. And, you know, are you averaging operating 13

points, or are you having multiple indexes? The thing 14

is -- 15

MS. IYAMA: So you'd have a separate value 16

of the index or of the metric for each of the 17

operating points. So that table that Tim had up on 18

his slide, that was the same fan, the performance 19

table of one fan with different ratings at each of the 20

represented operating points. 21

MS. ARMSTRONG: Yeah. So something that's a 22

little unique to what -- for those that aren't that 23

familiar with what the AMCA white paper -- kind of 24

approach they took. Something that's unique in what 25

318


they've done and is a little different from most 1

typical DOE regulations is instead of picking a rating 2

point, which you guys are all accustomed to, even 3

though you recognize your equipment operates over a 4

range of different points and a few operating points 5

in the field, they're rating at every operating point. 6

And so they're defining an operating range. I'm 7

sorry. And that whole range is what ultimately they 8

are -- I mean, as the approach would get administered, 9

that range would be certified to the department. 10

So instead of it being a single point for a 11

fan, they're going to be defining what this range of 12

conditions would be. Now there will be some bounds 13

around that range that we'll get into as we talk 14

through the test procedure and how you come up with 15

ratings, but I think the scheme is completely 16

different in something that's -- it's different in 17

terms of regulatory approach than the department 18

typically takes. And it's not just a single rating 19

point. It is a range reflective of the entire 20

operation range that the manufacturer wants to self-21

declare and sell its fans to be used in. 22

I saw some puzzled faces. It's different. 23

MR. SMILEY: It's normal for the industry -- 24

this is Bill Smiley, Trane. It's just normal for the 25

319


industry that we operate in. 1

MS. ARMSTRONG: Correct, which is why we've 2

kind of embraced that. 3

MR. SMILEY: It's not an issue for us. 4


MR. SMILEY: It's a DOE problem or change. 6

(Laughter.) 7

MS. ARMSTRONG: We implemented it in the 8

NODA, yeah. So, if you read the second NODA, we 9

implement it. We ran analysis on that exact 10

regulatory approach. 11

MALE VOICE: Well, I haven't had four years 12

to study all this stuff. 13

MS. ARMSTRONG: You're a fan manufacturer. 14

Anyhow, so that's that. I just wanted to -- 15

for those that may not be familiar, it's different, 16

but it's not something DOE is I would say embracing. 17

MS. JAKOBS: So this is Diane Jakobs. That 18

chart that we just looked at, it kind of looks like an 19

airflow chart for -- 20

MS. ARMSTRONG: Correct. 21

MS. JAKOBS: So we would have an FEI at 22

every -- 23

MS. ARMSTRONG: Point. 24

MS. JAKOBS: -- point. 25

320


MS. ARMSTRONG: So your table, your cert 1

table would look like this. 2

MS. IYAMA: We have pressure also in there. 3


MR. SMILEY: Yeah, statics -- Bill Smiley, 5

Trane. Statics across the top, CFMs down the sides, 6

like our normal. It's just instead of just having RPM 7

and brake horsepower like the industry has always 8

done, they've added a third column, which says or 9

whatever the metric is or -- I'm probably getting my 10

terminology screwed up here. But whatever the 11

measurement is, the ratio, FER, FEI, XYZ, whatever it 12

is, there would be an extra column for that. And the 13

methodology you're proposing or presenting here is 14

that I think you're suggesting it would need to be one 15

or above to be allowed legally to be used. 16

MS. ARMSTRONG: We haven't gotten there. 17

MR. SMILEY: Or something like that. 18

MS. ARMSTRONG: I mean, that's going to be 19

for like -- hopefully we get there in June, though. 20

MR. WHITWELL: Bob Whitwell from Carrier. I 21

think there is an important difference here, and that 22

is that what we publish today is based on what the 23

entire unit internal static pressure plus external 24

static pressure is. So it's an operating map for the 25

321


application in the building. 1

I think what this is talking about is for 2

the fan itself. So we don't have that data today. We 3

don't publish that today. So that's something that's 4

completely different than what we provide today. 5

MR. SMILEY: Yeah. Bill Smiley, Trane. 6

That's because the customer doesn't directly care what 7

the internal pressure losses of the unit are. He only 8

cares about what the external pressure capability is 9

for his application, what the cost, what the motor 10

size, the power, and all that are. So, if you base a 11

regulation on total fan static pressure or total fan 12

total pressure, we don't know what that is a lot of 13

times. 14

MR. SMITH: Well, this is Wade. The fan 15

that goes into your unit will under a regulated scheme 16

have a compliant operating range, and your unit, since 17

you don't catalog the fan performance, you catalog the 18

unit's air performance, the unit will have a 19

corresponding compliant operating range. And you do 20

catalog and you do know what the air performance of 21

the unit is. You catalog it. So now you just have to 22

impose upon what you catalog the compliant operating 23

range of the fan and the corresponding FER. 24

If you have the performance of the fan and 25

322


you have the performance, the air performance of the 1

unit, which you clearly do have, then you can 2

translate from one to the other. 3

MR. FLY: This is Mark Fly at AAON. So just 4

trying to get my head around this. As we looked at 5

say a unit performance chart, I would have the same 6

CFM because the CFM doesn't change. But the static 7

pressure available external would be less than the 8

static pressure of the raw fan and the minimum 9

testable configuration. And so you couldn't directly 10

look -- the table for the fan and the table for the 11

unit would not line up on the static pressure side. 12


MR. FLY: It would be incorporated, but the 14

FEI would be of the fan in the minimum testable 15

configuration. 16

MS. IYAMA: All right. So, in the next few 17

slides, I'm just going to go into a little bit more 18

details in the fan efficiency equation that's used to 19

establish the maximum allowable shaft input power and 20

then, based on that, the maximum allowable electrical 21

input power to the fan. So we went over what those 22

constants mean, so I'm just going to go through this 23

really quickly. 24

And there were some questions about, you 25

323


know, how did you get to the value of Q-zero and P-1

zero by matching the diameter effect of the FEG 2

curves, kind of mimicking flow and diameter. This is 3

just what those functions look like, like if you plot 4

a function that's P divided by P plus P-zero. And 5

then just to illustrate how that shape would change if 6

we increased P-zero a little bit or decreased P-zero a 7

little bit. So that's on the left graph. And then we 8

did the same for Q-zero. 9

Also, based on the data that AMCA had 10

provided to us, we did our own kind of curve fitting 11

to see how far or how close we'd get to those 215 0.4 12

values. We ended up pretty close, with a 234 and 0.4, 13

but looking at how sensitive the curves are to those 14

values, you know. 15

Also, if we get to presenting the NODA 16

results at some point, we could show you kind of the 17

impact of using a slightly higher or slightly lower 18

value for those constants. 19

MS. ARMSTRONG: Right. So I think just the 20

moral of this slide is that those constants are 21

representative of the data set we used to come up with 22

the NODA. We did a similar analysis that AMCA did 23

just for purposes of a) due diligence, but b) to see 24

how sensitive changes would be and changes in data 25

324


sets would be to those constants. And I think what 1

Sanaee is trying to say is that we didn't find that 2

the constants were overly sensitive, but I don't think 3

we're asking you to say these are the right constants 4

right now because, you know, as we move along in this 5

process and we add data to the data set and we revise 6

analyses, those constants may very well change. 7

We just want you to understand that the 8

constants didn't seem overly sensitive to data sets. 9

We are committed to updating them as we go through the 10

process, and we include more data, and the general 11

form of the equation on the next slide is going to 12

look something like this. We're there. Right. 13

MS. IYAMA: So another -- 14

MS. ARMSTRONG: So, in other words, don't 15

worry about the numbers quite yet. What we're trying 16

to talk through at a high level is the form of the 17

equation and the different components that it is 18

accounting for to get people onboard with the concepts 19

before we talk details of the -- when we get into 20

details of the analysis and how the actual data and 21

what the results say, that's when we need to worry 22

about the specifics. 23

So you can keep going. 24

MS. IYAMA: Yeah. So one point where the 25

325


NODA was different from the AMCA white paper is using 1

static efficiency -- well, an equation expressing 2

static efficiency as a function of static pressure and 3

flow to calculate the maximum allowable BHP for 4

unducted fans. And here I just have two slides to 5

kind of illustrate what the difference is if we use 6

one option or the other. It's just to kind of trigger 7

a discussion. 8

So for the ducted fans, where AMCA is using 9

an expression of total efficiency as a function of 10

flow and total pressure, we end up being able to 11

calculate the shaft input power based on flow and 12

total pressure. So that's your input data to the 13

function, which means that if two fans have the same 14

total pressure and same flow, they would get the same 15

sort of baseline maximum allowable shaft input power. 16

And then if you go on to the proposal or 17

recommendation that they have for the unducted fans, 18

where it's a metric based on static pressure, then 19

you'd have maximum shaft input power expressed as a 20

function of flow and static pressure, which means that 21

if two fans have the same static pressure and the same 22

flow, then they would be given the same sort of 23

allowance. 24

So what it looks for two theoretical fans, 25

326


fan A, fan B. Here it's just an example. They have 1

the same static pressure. So, if we're using option 2

1, these -- I just used a target of 62. It's just for 3

purely illustrative purposes, and then a slightly 4

lower target if we're talking about static efficiency, 5

so 0.58. 6

And the two other bullet points are sort of 7

a summary of the comments we got on that issue from 8

the first NODA, you know, concerns about using total 9

pressure because that would require defining the 10

outlet area, and that's not always straightforward for 11

unducted fans. And then other stakeholders saying 12

that, you know, when you use static efficiency, you 13

may be able to slightly change the way you define your 14

outlet areas also and sort of manipulate that to kind 15

of show that you have a higher efficiency than the 16

actual. 17

So, again, these are just to trigger some 18

discussions. I don't have any further slides on that 19

issue. 20

MS. ARMSTRONG: You guys have been waiting 21

to talk static/total, so here would be your chance. 22

I'm not sure who wants to take the lead here. 23

MR. SMITH: Well, let me just introduce it. 24

This is Wade Smith from AMCA. Most of the time, not 25

327


all, but most of the time, when a fan has no duct on 1

its discharge, the velocity pressure doesn't matter. 2

The velocity -- the energy that is imparted to the air 3

to speed it up, so to speak, is lost and of no 4

consequence. In other words, if you made it go 5

faster, you wouldn't bring the customer any benefit. 6

So there are real exceptions to this, and 7

that's why I said most of the time, because you have 8

jet fans where velocity is the only thing that 9

matters, induced flow fans, where it matters. There 10

are other examples where velocity matters. But 11

generally speaking, fans that have a free discharge to 12

the atmosphere, velocity pressure is of no consequence 13

to the application and of no value to the application. 14

And there's sort of a theoretical argument that says 15

therefore it doesn't belong in the efficiency equation 16

as a value add. It is not a value add. It's a waste. 17

But then there's the practical side of 18

static versus total, and the practical side is that if 19

you use static efficiency as the metric, then the 20

manufacturer can extract benefit from a -- what do you 21

call it, diffuser? What do you call those things? 22

Ellipse, evas, évasé -- on the outlet -- 23

MS. ARMSTRONG: Évasé. 24

MR. SMITH: -- to boost the energy -- to 25

328


boost the energy efficiency of the application, and it 1

provides a real benefit. 2

If you base the regulation on total 3

efficiency, évasé doesn't change the total efficiency, 4

but it does change the static efficiency, which is of 5

benefit. 6

So there are lots of arguments, and this is 7

another debate that lasted for years literally and has 8

not been -- you know, it isn't over. We still have 9

members who are on both sides of this question. But 10

the consensus is to use static efficiency on fans that 11

don't have ducts on their outlet. 12

I should just add velocity pressure on a 13

ducted application is of value because between the fan 14

discharge and the ultimate air discharge to the 15

occupied space if it's a ducted system or wherever the 16

duct ends, that velocity pressure is converted to 17

static, and it is used and exploited, knowingly or 18

otherwise, to overcome losses in the duct system. In 19

other words, velocity pressure has real value 20

delivered to the system because it is recovered in the 21

form of static pressure. 22

So I'll turn it over to folks who are a lot 23

more expert and have been participants in the debate. 24

I've been an observer. So, Tim, do you want to -- 25

329


MR. MATHSON: Tim Mathson from Greenheck. 1

Sanaee, can we put up a couple of the other slides 2

that were on the desktop there? Yeah. 3

MS. IYAMA: It's called static versus total. 4

MR. MATHSON: Static versus total. So just 5

to walk through the discussion a little bit, we insist 6

on using both of these, static for nonducted fans and 7

total for ducted fans, so that the metric correlates 8

to the energy consumed, for one reason, and to 9

encourage the right behavior and obtain the right 10

results. 11

If you go to the next slide, there is a 12

couple of misconceptions that I think that I know, you 13

know, why they occur. One is that fan static pressure 14

and efficiency is calculated from total pressure and 15

efficiency. And I know that our standard AMCA 210 and 16

5801 look like that because of the way they're 17

written. But if you look deeper into those standards, 18

the fan static pressure is what is measured during a 19

test, and then the velocity, the average velocity 20

pressure, is added on to that to get fan total 21

pressure. And then at the end, we say in order to get 22

fan static pressure you subtract it off again. 23

So there's a misconception that total 24

pressure is actually measured, but it's not. It isn't 25

330


during that test. In the static -- so the fan static 1

pressure and the fan static efficiency can just be 2

calculated directly from the measured values. 3

Secondly, I think people get twisted around 4

a little bit and think if we're not using total 5

efficiency we'll be ignoring energy that goes into the 6

accelerating of the air. And it's just kind of 7

getting it mixed up a little bit. Actually, the 8

opposite is true. If we use total efficiency for a 9

nonducted fan, we're accrediting fans for energy that 10

doesn't get used, the velocity pressure at the 11

discharge of the fan. And so the difference -- the 12

best example I think is just a sidewall prop fan, 13

small versus large, can develop the same total 14

pressure, can operate at the same total efficiency. 15

But the static pressure would not be as high for that 16

small fan because such a large portion is in velocity 17

pressure. 18

What's the next one? Oh, arguments. So 19

we've argued about this within AMCA because there are 20

some that I'm going to say didn't understand it. But 21

anyways, we've got a couple of pages about fan 22

velocity pressure in a nonducted fan as being not 23

useful energy. And other people -- Wade Smith has 24

said that if we use total efficiency for fans without 25

331


outlet ducts, it gives an advantage to smaller fans, 1

and we don't want to do that. We want to save energy. 2

Mark Steven's words that he -- he talks 3

about the industry is not willing to pay for that 4

velocity pressure in fans. And the European 5

regulation also understand that and uses a dual -- 6

well, use static and total pressure, but it's not, 7

like I say in these notes, it's not a dual metric. 8

It's just a different measurement applied to two 9

different fan types. 10

And, you know, I think one of the 11

purposes -- we don't talk about a goal, but a goal 12

with any metric is to have a correlation with the 13

energy consumed. In the next slide, and I don't think 14

this is too controversial, but in the next slide, you 15

know, we're proposing a metric based on how the fan is 16

tested. It would be ideal if we knew how the fan was 17

applied, but we don't. And so we're trying to reflect 18

that. And this is just looking at those two different 19

cases, ducted and nonducted, and how well does the 20

testing correlate with how they're applied. 21

If the fan is tested nonducted and applied 22

nonducted, obviously the two correlate. Or if it's 23

tested ducted and applied ducted, there's correlation 24

there also. The bottom left-hand box, if a fan was 25

332


tested nonducted, it will never be applied in a ducted 1

situation. So that's kind of, that's a bold term 2

there, never. And the reason I say that is because if 3

we have any means to apply a duct to a fan, we'll test 4

it that way because we get better results. And so the 5

only fans that we test nonducted are those that don't 6

have a flange or a defined outlet where we can put a 7

duct on. 8

So the bottom left doesn't exist. The top 9

right are the ones where we don't quite match up. So, 10

if we test a fan that's ducted and it's applied 11

nonducted -- and I listed two major ones, double-wide 12

fans and air handlers that are blow-through, so 13

they're not on the end of the air handler. Those 14

would be almost always tested as a ducted outlet 15

because they might be applied as a ducted outlet, and 16

yet they're quite often applied nonducted. 17

And then utility fans that you may find on a 18

roof that's just an exhaust fan that's fairly common 19

in a -- which would be the responsibility of the 20

contractor to provide an outlet duct, which oftentimes 21

they don't. 22

So those are two major ones, and we could 23

think of a lot of smaller ones, niches. But I guess 24

that's kind of a point of discussion here is, does 25

333


that require any more -- any more consideration for 1

that corner, that quadrant, the top-right quadrant 2

there. 3

MR. FLY: This is Mark Fly with AAON. When 4

you say ducted, you're meaning primarily discharge? 5

MR. MATHSON: Only discharge, yes. 6

MR. FLY: Only discharge. 7

MR. MATHSON: It doesn't -- 8

MR. FLY: Even though the fan performance 9

will change if you duct the inlet or don't duct the 10

inlet. 11

MR. MATHSON: Right. The fan performance 12

won't change appreciably if you duct the inlet or the 13

outlet. I'm sorry. If you duct the inlet. 14

MR. WOLF: This is Mike Wolf. Let me maybe 15

phrase that a different way, Mark. For a ducted -- I 16

mean, first of all, I think the answer is yes. It's 17

just for ducted outlet, okay? But for a ducted inlet, 18

for your ducted inlet, you're going to take that into 19

account in your external static pressure calculation 20

for selecting the fan, I think, usually. Am I wrong 21

there? At least that's been my experience. 22

MR. MATHSON: Well, this is Tim again. As 23

far as an AMCA fan test goes, whether we put a duct on 24

the inlet or we put an inlet bell on the fan on the 25

334


inlet, we get about the same performance, and the bell 1

just represents a no-loss entry into the duct. Well, 2

different -- yeah, different fans will behave slightly 3

differently. 4

MR. HARTLEIN: Yeah. This is Dan Hartlein, 5

TCF. I was just commenting that the inlet bell quite 6

often is actually better than the duct, from an inlet 7

configuration perspective. So there may be more loss 8

in the duct than there is actually in the inlet valve 9

as a rule. 10

Is there any -- a question for me would be 11

is there any controversy to this -- to our -- because 12

we're almost discussing it like we're trying to sell 13

something here, and it sounds to me in the room it 14

feels like everybody agrees that this is good. 15

MS. JAKOBS: Well, this is Diane from -- 16

MR. HARTLEIN: No, we're not there yet, 17

Diane? 18

MS. JAKOBS: -- Rheem. 19

MR. HARTLEIN: Okay. 20

MS. JAKOBS: So are you saying that upper 21

right-hand box is a loophole so that the -- 22

MR. MATHSON: Well, this is Tim. I would 23

say it's the type of fan where the metric, which would 24

be based on how the fan is tested, wouldn't reflect 25

335


necessarily the energy consumed as applied. But in 1

those two cases that I put there, they're very close. 2

Those two cases are not a problem. The bigger 3

problem would be is if we tested a fan or had a metric 4

in total efficiency and it was applied nonducted. I 5

mean it was -- I'm sorry. If there was a significant 6

velocity pressure component, like a propeller fan. 7

MR. HARTLEIN: Diane, if I could. This is 8

Dan, TCF. I think in my experience that square is 9

limited where we've actually tested a ducted fan 10

that's been applied in a nonducted configuration. 11

There are some pretty unique applications when that 12

happens. I'm thinking of a device we build that 13

actually dries the greens for a golf course, for 14

example, okay? So they use a centrifugal fan to do 15

that. They put on a nozzle on it. Then they're 16

trying to blow as much high-velocity air across, so 17

they're converting a fan to be a velocity machine in 18

that case. Perhaps it's not a fan in that case. It's 19

doing something different. 20

What's that? Yeah, we do also fans -- if 21

anybody in the room is a skier, the fans that you see 22

in the snow machines, those are also high-velocity 23

machines which are undoubtedly rated as a ducted fan 24

because typically that fan would be applied, it's the 25

336


anomaly that it's applied without a duct. 1

I had another thought as we went there. Oh, 2

dust suppression in a coal yard, for example, we will 3

use a very, very similar machine in order to blow mist 4

into a dusty environment because the mist will tend to 5

serve as a surfactant and pull the dust out of the air 6

in the industrial market. 7

So these are all very, very small and unique 8

and special applications. I would be hard-pressed -- 9

maybe somebody else in the room could -- to come up 10

with a large population of fans that are tested ducted 11

and then applied unducted. I think they're the more 12

the exception, not the rule, would be my thinking. 13

MR. SMILEY: Bill Smiley, Trane. I don't 14

necessarily agree with that because in a lot of 15

applied products, the fan is a housed centrifugal fan, 16

and it is blasted right into a blow-through situation, 17

just like he says on the slide. 18

MR. HARTLEIN: Yeah. But in that -- 19

MR. SMILEY: So there is no discharge on the 20

fan. 21

MR. HARTLEIN: In that case, there is a 22

system resistance as well. But you're right, yeah. 23

MR. SMILEY: But the other point I want to 24

make is if you look at the definition of total 25

337


pressure per AMCA, which is the measured static 1

pressure plus the velocity pressure calculated from a 2

discharge area, and you said, well, whether it's got a 3

duct or not, if it's unducted, the static pressure is 4

the total pressure based on that definition because 5

the area is infinite. 6

MR. BUSHNELL: Peter Bushnell with Carrier 7

again. I had a few comments on this. First on the 8

whole -- this chart here that we're looking at. I 9

think this is kind of consistent with what we were 10

talking about yesterday in the EU-327, categories A, 11

B, C, D. So they sort of have this already set, and I 12

think it's kind of fine the way they have it. 13

If you don't have an outlet duct, you should 14

be using static pressure and static efficiency very, 15

very clear. I don't think -- I mean, this goes way 16

beyond sort of just sort of this industry thing. This 17

is like basic turbo machinery. It's like in the 18

textbooks, you know, very basic stuff. 19

So there's no doubt that if you don't have 20

an outlet duct you should be talking about the static 21

efficiency and static pressure. And so why is that 22

really important? I think Tim was alluding to this 23

before and some of the other folks that commented. We 24

really want to have high static efficiency when we're 25

338


driving flow out to free discharge, and one of the 1

ways to achieve that is, as Tim mentioned, you can 2

have a -- just say this is an axial fan. You can have 3

a bigger axial fan. It will have a higher static 4

efficiency because you're basically not blowing all 5

that kinetic energy out through the outlet of the fan. 6

You know, you're reducing the discharged kinetic 7

energy loss. 8

The other way to deal with that is to put a 9

diffuser on the outlet and recover that kinetic energy 10

and transform it to static pressure. And these are 11

the kinds of things that DOE should be really 12

encouraging, is that, you know, we really want to seek 13

energy efficiency for fans where we have these kinds 14

of abrupt discharge conditions that are necessary in 15

many cases, like free discharge in axial fans on a 16

condensing system. Encouraging means for recovering 17

that energy, that discharged energy, and minimizing 18

the exit loss is a big opportunity. 19

And the previous comments about centrifugal 20

fans embedded in systems and abruptly discharging, 21

again, there's many cases where we need to do that in 22

equipment due to flow distribution. We need to blow 23

through or draw through heat exchangers that are in 24

close proximity to the fan, and we actually achieve 25

339


interaction effects that can even be favorable and 1

help the fan and the system net performance and energy 2

draw in those cases. But it's still like the static 3

pressure that matters in those cases, the static 4

efficiency, and when we achieve high static efficiency 5

in those embedded systems where you have abrupt 6

discharge, we're doing the right things when we drive 7

that static efficiency. 8

So, again, I think it's very clear. If you 9

look at the European regulation, they've got it 10

figured out. I think we can learn from that. I 11

wasn't on earlier. I don't know if you brought that 12

back up. We talked about it a little bit yesterday. 13

But I don't think this should be a controversial area. 14

I think this should be really pretty clear, static 15

efficiency, static pressure whenever we're discharging 16

abruptly and losing the energy at the exit. 17

MR. HARTLEIN: So can I -- this is Dan from 18

Twin City Fan. I agree 100 percent with what you 19

said. My question back to the DOE, when the NODA came 20

out, it went mostly or predominantly or all I guess to 21

total, didn't it? And so our question is these 22

arguments that have been made here today, it seems to 23

me we've got pretty good consensus around it, with the 24

exception of the NODA. So can you guys give us some 25

340


insight into what your thinking was? You've had some 1

good technical minds wrapping your heads around this 2

for a while. Where were you with total as opposed to 3

the total/static argument? 4

MS. ARMSTRONG: We're ready to vote. 5

MR. HARTLEIN: Ah, okay. 6

MS. ARMSTRONG: How about that? 7

MR. HARTLEIN: Great. Thank you. 8

MR. MATHSON: This is Tim from Greenheck. 9

Just one more comment. As long as it looks like 10

there's pretty good consensus here, the one -- well, 11

Dan mentioned a couple of exceptions, and I'll just 12

mention another one, jet tunnel fans. In the European 13

regulation, they will be -- so that's a fan that's 14

free discharge, but its purpose is to increase the 15

momentum of the air, and so that's not wasted energy 16

in that case. So what they are doing is they're 17

rating that jet tunnel fan in total efficiency for 18

that purpose. So there may be other exceptions to 19

that, like laboratory exhaust fans or something, so 20

just to keep that in mind. As a general rule, we're 21

talking about differentiating into these two different 22

ducted and nonducted categories, but there may be a 23

couple of exceptions. 24

MS. ARMSTRONG: Okay.. 25

341


MR. BUSHNELL: It's Peter again. I think 1

even though I was kind of firm about that whole thing 2

with, you know, the delineation, there are systems 3

where even though static pressure is really what we're 4

looking at, static efficiency like for say a 5

condensing or an air-cooled chiller, we do care about 6

the discharged kinetic energy because we need to get 7

the heat away from the system. 8

And so the optimal design point for the fans 9

in some of those systems may be not exactly right on 10

peak static efficiency or peak total. It's kind of 11

something that's a little bit special to really 12

optimize at the system level. I think there have been 13

some comments made about whether or not heat rejection 14

equipment should be potentially exempt from this, and 15

that's one of the -- that might be a factor, I think, 16

is that heat rejection equipment, you need to move the 17

heat away and use that kinetic energy to do that. 18

MR. SMILEY: This is Bill Smiley. To add on 19

to that comment, a typical nonducted application of an 20

air conditioning piece of equipment like a fan coil 21

needs to have a certain velocity leaving the unit so 22

that you get the throw and distribution of the cool 23

air or heated air or ventilation air, whichever of 24

those that you're after. 25

342


MR. BURDICK: This is Larry Burdick. I 1

would agree with those comments that you want to 2

remove the heated air maybe in our case away from the 3

unit to avoid recirculation, avoid upset situations or 4

severe upset situations in low ambient winds, or those 5

types of conditions. 6

MS. ARMSTRONG: After the not-so-7

controversial discussion, here is where we are. What 8

you've heard is the presentation of metric. It's been 9

a broad presentation. We would like to get a vote 10

before lunch, and this will be the first one. So 11

generally speaking, this would be the fan efficiency 12

equation. This is going to show static efficiency for 13

nonducted fans. It's going to show total pressure at 14

the operating point for ducted fans, or static 15

efficiency and total efficiency and then static 16

pressure and total pressure. So are we ready to vote, 17

yes or no? 18

MR. SMITH: What are we voting on? 19

MS. ARMSTRONG: That. So the constants have 20

to be determined and will be determined as we move 21

along in the analysis. But you're generally voting on 22

the form of the equation, the components of the 23

equation, and then the allotment of static versus 24

total for ducted versus nonducted, or vice versa I 25

343


guess is the way I just said it. 1

MR. SMILEY: Does it include the electrical 2

portion, this brake horsepower? 3


MALE VOICE: Use your microphone, please. 5

MR. SMILEY: Sorry. This is Bill Smiley. 6

MR. BUSHNELL: Before you vote on this, you 7

might want to be really clear on the definition of 8

static efficiency and total efficiency from a very 9

basic standpoint. These are the definitions of what 10

you're using for your metrics I guess, the standard. 11

But I'm not sure that DOE has a completely clear 12

indication of what the definition of fan static 13

efficiency is and total efficiency based on what I've 14

read. 15

MS. ARMSTRONG: It's on the slide. 16

MR. SMILEY: This is Bill Smiley of Trane. 17

It's on the slide. It's flow times pressure divided 18

by horsepower and a conversion factor. 19

MR. BUSHNELL: That's not the definition of 20

static efficiency -- 21

MR. SMILEY: What?. 22

MR. BUSHNELL: -- for a fan. 23

MR. SMILEY: I'm sorry. What is then? Can 24

you enlighten us? 25

344


MR. BUSHNELL: There's a definition for 1

static efficiency for a fan is that it's the pressure, 2

the static pressure, the product of the static 3

pressure times the flow rate divided by the shaft 4

input power. 5

MR. SMILEY: That's just what we said. We 6

said brake horsepower is shaft input power a little 7

bit ago. 8

MR. BUSHNELL: Okay. 9

MS. ARMSTRONG: So, Bob, I'm going to ask 10

you, are you in a position to vote, or do you need 11

lunch to talk with him? 12

MR. WHITWELL: I think I need lunch to talk 13

with Peter. 14

MS. ARMSTRONG: Because you guys are on the 15

same -- yep. 16

MS. WALTNER: This is Meg. I just, I have a 17

question probably for Sanaee. I just wanted to 18

understand, you know, did DOE have a different 19

rationale for using total pressure for everything in 20

the -- no. 21

MS. ARMSTRONG: Not necessarily. We kept 22

the same just for an equitable standpoint, but there 23

wasn't a strong opinion one way or the other, yeah. 24

Yeah. So we have one person -- Bob, we have 25

345


one person that's leaving early and needs -- it would 1

be a good idea if he could vote before lunch. Can you 2

resolve it in five minutes or less? 3

MALE VOICE: We can try. 4

MS. ARMSTRONG: Can you go try for five 5

minutes, please? 6

MR. WHITWELL: Hey. Hey, Peter? 7

MR. BUSHNELL: Yeah. 8

MR. WHITWELL: This is Bob. Can you just 9

text me quick the phone number that you're at so that 10

we can give you a call? I'll call you along with a 11

couple of the other manufacturers, okay? 12

MR. BUSHNELL: Sure. Okay. 13

MR. WHITWELL: Okay, thanks. Yeah. So you 14

can email it to me. I'll get it out of my phone, 15

okay? 16

MS. ARMSTRONG: So five minutes, we'll 17

reconvene to vote on the slide, yeah. 18

(Whereupon, a brief recess was taken.) 19

MS. ARMSTRONG: So, Sam, I'm going to ask 20

you can you provide a 60-second overview of what we're 21

voting on on this slide. 22

MR. JASINSKI: Sure. So what we're voting 23

on here is just the general form of the equation to 24

determine the target efficiency and based on that 25

346


target efficiency what the maximum allowable brake 1

horsepower would be at any given operating point. And 2

we've added in red text the modification that for 3

ducted fans P would be total pressure and efficiency 4

would be total efficiency, and for nonducted fans P 5

would be static pressure and the efficiency would be 6

static efficiency. 7

MS. ARMSTRONG: Okay. Can you guys call a 8

vote? 9

MR. BOSWELL: Okay. So a vote for, I guess 10

we're doing thumbs up, agreement, can live with; 11

thumbs down, not in agreement. 12

MS. ARMSTRONG: So, Joanna, you have to tell 13

us your vote because you're unmuted. 14

MS. MAUER: Do I say that my thumb is up? 15

MR. BOSWELL: Okay. Okay. So I -- 16

MS. ARMSTRONG: We'll get that. Vote first. 17

MR. BOSWELL: Okay. So I'm saying counting 18

thumbs down, I'm seeing one, two, three, four. So we 19

do not have a consensus. Our definition of consensus 20

is if we have three negative votes there is no 21

consensus, so I guess what I would -- 22

MS. ARMSTRONG: Four. 23

MR. BOSWELL: Four. 24

MR. FINE: And who are the four negative 25

347


votes besides that guy? 1

MR. BOSWELL: Okay. I'm sorry, I'm looking 2

at the wrong one. Somebody left the wrong stack of -- 3

okay. So it is four. 4

MR. FINE: We do have a rule, but if you 5

vote against you have to explain why. So why don't we 6

start with -- 7

MALE VOICE: I'll ask Bill to start. 8

MR. SMILEY: The reason I voted -- 9

MALE VOICE: Excuse me. I think you only 10

have three negative votes. 11

MS. ARMSTRONG: You have AHRI right behind 12

you. Karim? Karim is the alternate for Laura. 13

MR. BOSWELL: He's the alternate for Laura? 14


MR. FINE: Why don't we start with you, 16

Bill, and you can say why -- 17


There's a couple of issues I have, and I hope I can 19

resolve them. One is I'd like to see a singular 20

definition of what efficiency is, as like a line above 21

that equation, that first equation. What is the 22

definition of efficiency? And the second thing is -- 23

MS. ARMSTRONG: I don't understand that. 24

What are you asking for? Like what is the metric 25

348


going to be? 1

MR. SMILEY: I want to see a singular 2

definition of what static efficiency and total 3

efficiency are without that middle stuff in there. I 4

want to see it -- 5

MR. JASINSKI: What's the middle stuff 6

you're talking about? 7

MR. SMILEY: There's an equation there with 8

a left-hand side of the equal sign, there's a part 9

that's in between the two equal signs, and then 10

there's a third part to the right. 11

MR. JASINSKI: You want a definition for 12

this term here? 13

MR. SMILEY: Yes. 14

MR. JASINSKI: And this is not sufficient? 15

I'm saying you want it in -- 16

MR. SMILEY: I want a singular definition. 17

MS. ARMSTRONG: He wants either the middle 18

or the right, not both. 19

MR. SMILEY: Yes. 20

MR. JASINSKI: I see. 21

MS. ARMSTRONG: In other words, he's 22

asserting they're not equivalent. 23

MR. SMILEY: I don't know. 24

MS. ARMSTRONG: Okay. 25

349


MR. SMILEY: I don't know if they're 1

equivalent. I suspect that they may be, but for the 2

definition, I want a singular definition, not a dual 3

definition, okay? 4

MS. IYAMA: This is Sanaee. Would you be 5

comfortable if we deleted this? 6

MS. ARMSTRONG: Yeah, just take that out. 7

MS. IYAMA: And then this? 8

MR. SMILEY: No. I want to see what the 9

official technical definition of efficiency is. And 10

if you are saying that the definition of efficiency is 11

some bogus or some bogey factor target -- excuse me. 12

I didn't mean to say bogus. Some bogey factor target 13

efficiency times Q over P divided by the quantities of 14

Q+Q times P+P, that is not the industry accepted 15

definition of what efficiency is currently. What I 16

want to see is what that definition is. It's a simple 17

equation you put right above that. 18

Now it may in the final, a month or two from 19

now, end up being exactly what that is, but I cannot 20

support that form of a definition of efficiency until 21

I understand how it affects the products of the 22

company that I represent. 23

I have not had four years to do all this 24

analysis and study all this data. I appreciate all 25

350


the work that's been done to get to this point and I 1

think there's a lot of smart people doing a lot of 2

highly technical stuff to get here, the data to back 3

it. But my industry, my company has not had a chance 4

that I'm aware of to evaluate this, so I cannot in a 5

true sense vote for this until I understand what the 6

implication is. There may be a better way. I don't 7

know. That's why I'm voting no. 8

MR. JASINSKI: Can I ask a clarifying 9

question? So it sounded like you were trying to 10

evaluate whether efficiency STD is equivalent to the 11

industry accepted efficiency definition. 12

I think the answer is no, but not because 13

it's not equivalent and it should be, but the target, 14

efficiency target would be equivalent to the industry 15

defined definition of efficiency, static efficiency or 16

total efficiency, and the efficiency STD basically is 17

modified by that second term in that equation to be 18

de-rated with flow and pressure to account for the 19

inherent difficulty in achieving higher efficiencies 20

at lower airflows and lower pressures. 21

Am I right in that, Wade? 22

So all I'm trying to clarify is that for 23

your exercise the comparison should be between -- the 24

intent is that efficiency target is equivalent to the 25

351


industry definitions that you're talking about, not 1

the efficiency STD. 2

MALE VOICE: Make that clear. 3

MR. JASINSKI: Okay. 4

MR. SMILEY: Bill Smiley, Trane. If you 5

would make that clear -- 6

MS. ARMSTRONG: You can suggest stuff. 7

MR. SMILEY: If you would make that clear, 8

that would be very good. 9


MR. SMILEY: Ashley, you had a comment? 11

MS. ARMSTRONG: I said you can make 12

suggestions. 13

MR. SMILEY: I just did. That's a 14

suggestion. I still cannot vote for that form of the 15

equation that you have up there because I don't 16

understand how that interacts with the products that 17

my company manufactures. 18

MS. ARMSTRONG: But that's like saying -- 19

MR. SMILEY: I have not had a chance to 20

evaluate it. I don't know if that's the right form. 21

It might be there should be a squared function in 22

there. 23

MS. ARMSTRONG: So can I ask a question 24

then? 25

352


MR. SMILEY: It might be a log function in 1

there. I don't know. 2

MS. ARMSTRONG: Are you taking the position 3

then at the table that you're not going to vote on 4

anything until you determine if you agree to the 5

standard level at the very end? Because that's what 6

you're saying. Until you can agree to how the 7

standard impacts your company, are you not going to be 8

voting on anything? Because I think the committee 9

should know that, because we can't make progress that 10

way. Is that your position? 11


That is not my position. My position is I need time 13

to evaluate, understand, and analyze what all this is 14

going to do to my company so that I can make an 15

intelligent vote on what I think is right and what we 16

think is right. You're asking for us to -- 17

MS. ARMSTRONG: We're not voting on the 18

targets, right? We're voting on the form of the 19

equation. 20

MR. SMILEY: I know, I know we're voting on 21

the form of the equation. I just told you I don't 22

know that I would agree that that's the correct and 23

only form the equation could ever be in. I don't 24

know. Maybe there's other function that would be 25

353


better. 1

MR. MATHSON: A comment. This is Tim 2

Mathson from Greenheck. I can understand there's some 3

confusion because there's several steps that are 4

included in one line here, and I think if we just 5

separated those out step by step it would be more 6

clear as to what we're voting on. 7

MR. FINE: Well, at this point, just as sort 8

of a procedural way to end this, I think there should 9

be some more discussion on the point that Bill's 10

raising. It would be valuable to talk about it more. 11

But maybe we could first go and let everybody be 12

heard who had voted against it. Maybe they'll have 13

some of the same common concerns, and then we can deal 14

with the concerns and see if we can't, how do you want 15

to say it -- 16

MS. ARMSTRONG: Okay. That's okay. Do you 17

want to? I mean, I think at this point, you know, we 18

go to lunch. I mean, that's where we are. This is 19

the most simplistic thing we need to wrap our hands 20

around in order to start making progress with metric 21

and then test procedure analysis so you can actually 22

talk about levels and see how things might impact your 23

company. You know, this is how we have to move 24

forward. 25

354


We can try to break down this into 1

components. Actually, why don't you guys try and 2

break this down into components as to what you would 3

like to see and bring it back to the group for 4

discussion after lunch so we can move forward. I 5

think that's reasonable. 6

MR. BOSWELL: The other thing that I would 7

say is, you know, the whole idea with trying to reach 8

consensus as we go along is to kind of build towards a 9

final term sheet. Ultimately there's a vote on the 10

final term sheet so that if other complications come 11

up that people didn't anticipate, I don't think that a 12

vote on any particular matter forecloses that. It's 13

just a way of trying to build towards seeing if we can 14

reach consensus on a term sheet. 15

MR. FINE: So why don't we take a break for 16

lunch and come back and -- 17

MR. BOSWELL: It is now about 10 after 12. 18

I presume some people are going to want to talk during 19

lunch. And we're breaking at 3:00 today as I recall. 20

Okay. So we will return at 1:00. 21

(Whereupon, at 12:09 p.m., the meeting in 22

the above-entitled matter was recessed, to reconvene 23

at 1:00 p.m. this same day, Tuesday, May 19, 2015.) 24

//25

355


A F T E R N O O N S E S S I O N 1

(1:14 p.m.) 2

MR. JASINSKI: Okay. So, to pick up where 3

we left off, I think we want to resume the post-voting 4

discussions. We're going to allow all the "no" votes 5

to provide their explanation for why they're voting 6

no, give them an opportunity to present any 7

alternatives if they have them, and if they don't have 8

an alternative ready to be presented, to explain a 9

path forward in terms of what they're willing to do to 10

get to an alternative and on what time frame. 11

So we'll just kind of go through that 12

methodically, give each of them a chance to make those 13

statements, and then after that I think what we will 14

do is just continue to show what the department has 15

done in terms of metric, test procedure, et cetera, 16

just so that we can get feedback on those things that 17

have already been done and inform everyone else of 18

what those things are. 19

So, Bill, I think we left off with you. You 20

just handed me a note to try to clarify some of the 21

changes that you were suggesting. But why don't we 22

start over now that everybody's food coma is starting 23

and we'll just, you know, explain why you voted no, 24

any alternative, and if you don't have an alternative 25

356


what the path forward to get an alternative would be. 1

MR. SMILEY: Okay. Bill Smiley, Trane. The 2

addition I asked for to be added to this slide was the 3

singular definition of what efficiency is, and I gave 4

you on a piece of paper what that efficiency would be 5

for both static and total. The difference would just 6

be the pressure. 7

MR. JASINSKI: Right. I have the static one 8

here, but the changes would be that the fan efficiency 9

would be total fan efficiency and the pressure would 10

be total pressure. 11

Is this big enough for everyone? I can try 12

to make it a little bit bigger. 13

MR. SMILEY: Now the second comment. In 14

general, I agree that an equation that compensates for 15

operating a point on a fan and a fan selection that is 16

based on performance with some adjustment for flow and 17

pressure similar to what's shown up there is a good 18

thing. I appreciate all the work that AMCA and DOE 19

and you guys have done to develop this metric. The 20

only problem I have is the form of that equation. I 21

don't know if that's the optimum form of the equation 22

based on the products that my company makes. I was 23

told that somebody from my company participated in 24

this and I will have to check with them and get a data 25

357


dump from them. 1

In general, I agree with the concept of 2

having an equation similar to this. What I don't know 3

yet, because I have no data to back it or no 4

knowledge, is should that be exactly the form where 5

you add a little bit of CFM and you add a little bit 6

of pressure to come up with the relationship. That's 7

what I was voting no against. I was understanding 8

that if I voted yes that would be written in stone and 9

could never be changed. If that is not true, then 10

restate what we're voting on, I may change my vote. I 11

don't know. I don't know the procedures and the 12

process here, but that's basically my comment. 13

MR. JASINSKI: Okay. We were voting on the 14

general form, so what I would -- but it's good to know 15

that you are in general agreement with the concept. 16

So I think in keeping with the ground rules we would 17

just ask what is -- so you don't have an alternative 18

immediately. What can be done to get that 19

alternative? What can you do to get that alternative 20

and what's the time frame to do it? 21

MR. SMILEY: I would say by the end of the 22

week if he's available I can consult the other person 23

within the company that did a considerable amount of 24

work on this. 25

358



MR. SMILEY: And I would defer until I get 2

to that point. I make an assumption now that he 3

considered all products in scope and that his data 4

dump and opinion is valid, but I would reserve that if 5

he did not consider all potential products that I may 6

request a little more time to evaluate that. 7


MR. SMILEY: But again, in concept, I'm 9

agreeing with this type of methodology. I'm not 10

trying to stall anything. I'm not trying to be a 11

roadblock. And I know some people's perception is 12

we've had a long time to think about this and to look 13

at it and work on it, but to tell you the truth, I 14

haven't. 15

MR. JASINSKI: Okay. So I think that's a 16

good example of exactly what we want to hear in terms 17

of making progress. So for each of the other "no" 18

votes, I would say if you follow, if you do exactly 19

what Bill just did, I think that's what will be most 20

helpful. So I don't know who wants to go next. 21

Larry, did you want to weigh in? 22

MR. BURDICK: I have a message that they 23

can't hear on the webinar. 24

MS. ARMSTRONG: We're working on it. 25

359



MS. ARMSTRONG: We have to get our audio 2

folks in here. 3

MR. BURDICK: Okay. So Larry with SPX. One 4

of the things that I would be interested in is seeing, 5

you know, what the further development of this is. 6

You know, there's been two different NODAs published, 7

have evaluated, you know, the one that's labeled NODA 8

2, was not, you know, necessarily certain that it was 9

working properly or what all that consisted of. You 10

know, we'd like to see, I think I'd like to see a full 11

explanation of that, you know, as part of this. 12

In general, I'm in agreement with, you know, 13

this page, but I don't know how it affects, you know, 14

things downstream. 15

MR. JASINSKI: Okay. So I think we're 16

planning to get into -- 17

MS. ARMSTRONG: I was going to say I think 18

at our next meeting our plan is to walk through the 19

NODA analysis and those spreadsheets, so hopefully 20

that will give you the tools necessary to come to an 21

opinion on this one. 22

MR. BURDICK: Yeah. And, Ashley, I would 23

request that we have that available to everyone, you 24

know, well before the meeting so that we can perform 25

360


what if's and other scenarios. 1

MS. ARMSTRONG: Tools? 2

MR. BURDICK: Yes. 3

MS. ARMSTRONG: So they're available now and 4

I've downloaded them out of our docket and I can get 5

them working. So, if you've downloaded a version 6

that's not working, let me know after the meeting so 7

we can fix that. 8

MR. BURDICK: Fair enough. 9

MS. ARMSTRONG: Because they should be able 10

to be working now, the same ones that have been in the 11

docket. 12

MR. JASINSKI: Okay. I think that's two out 13

of the four. Do we have -- go for it. 14

MR. AMRANE: This is Karim with AHRI. I 15

think I consider the same as Bill. I think with the 16

addition here, I think we're fine with that. 17

Regarding a vote, I think it would be good 18

if we decide to vote on this slide that at least this 19

vote should be contingent upon having the possibility 20

of modifying this equation. If there's new data that 21

shows that we can modify the equation in the future or 22

if you want to give us more time to come back, you 23

know, at the next meeting with a definite yes or no. 24

It's up to you how you want to proceed. 25

361


MR. JASINSKI: Okay. Next meeting. We'll 1

take care of it on the next meeting. There was one 2

more. 3

MR. ROY: Aniruddh Roy, Goodman. I share 4

some of the same concerns as Bill and Karim mentioned 5

and they've been addressed through the conversation, 6

so no further comments. 7

MR. JASINSKI: Okay. Thank you. 8

So I think at this point we will just 9

continue with a little bit more information in terms 10

of the test procedure. I think it's going to come in 11

the form of just kind of a list of some information 12

that we're seeking. 13

MS. IYAMA: All right. So we've got two 14

sets of slides. The first one is sort of an overview 15

of the default values that were mentioned earlier to 16

get to the wire-to-air metric. Let's see. Okay. So 17

let's see if I can open the previous presentation. 18

(Pause.) 19

MS. IYAMA: Okay. So earlier we saw this 20

concept of adding default values to get to wire-to-air 21

metric, although we would be testing for shaft input 22

power, so that's when establishing the consumption of 23

the fan. 24

And then here, that energy consumption of 25

362


the fan would be compared to the standard level, so 1

based on that fan efficiency equation that Sam talked 2

about, combined with default values. Default values 3

for the motor, transmissions, and then for the fan, if 4

it's sold with controls, default values for the 5

controls. So that's what we're going to go through 6

here. 7

So for default values for the motor, when 8

calculating what in the NODA is called FER standard, 9

which would be the maximum allowable electrical input 10

power of the fan, we used motor efficiency values that 11

were at the level of the upcoming standard for medium 12

electric motors. That's sort of a description of what 13

the scope of that regulation is, the medium electric 14

motor's regulation. Basically it's three phase AC 15

induction motors. 16

Some of the exclusions which are pretty 17

important when we're looking at fan is that this 18

regulation doesn't cover totally enclosed air over 19

motors which are often used to drive fans. 20

MR. SMILEY: Excuse me. This is Bill 21

Smiley, Trane. What regulation are you talking about 22

that doesn't cover the TEAO motors? 23

MS. IYAMA: The medium electric motors. 24

MR. SMILEY: Oh, okay. The NEMA code. 25

363


MS. IYAMA: The DOE regulation. 1

MR. SMILEY: So is this a new DOE standard 2

that's coming out? 3

MS. IYAMA: So the final rule has been 4

published and I think it's going to be starting in 5

2016 if I'm not mistaken. 6

MR. SMILEY: So a new motor efficiency 7

standard covering what motors again? 8

MS. IYAMA: These motors. 9

MR. SMILEY: So this is an update to the 10

prior standard? 11

MS. IYAMA: Yup. 12

MR. SMILEY: So we'll be having new motors 13

that we'll have to qualify for all our equipment 14

presumably. 15

MS. IYAMA: If it's motors in the scope of 16

that regulation which are described on here, which 17

basically it's AC induction, three phase motors, with 18

a few exceptions that are listed below. And I think 19

for fans the most significant one is the TEAO motors. 20

So, you know, the idea is that those default 21

values are going to be representing sort of the 22

conservative side of the efficiency of a motor found 23

today on the market, and so here in the NODA and in 24

the calculation of the metric these are the values we 25

364


used. 1

I'm also going to present in the next slide 2

some default values for TEAO motors. These were not 3

used in the NODA. It's just to show one way that we 4

could do this and get feedback from the working group. 5

MR. HAUER: Sanaee, it's Armin Hauer of ebm-6

papst. Are you using minimum efficiencies or nominal 7

efficiencies according to the regulation? 8

MS. IYAMA: So I'm going to go through those 9

values in the next slide. So for motors that are AC 10

three phase and regulated motors, we would be using 11

those default values which are here. And again, 12

that's up for discussion, but what we have in the 13

first table is the table that comes directly from the 14

CFR, the medium electric motor regulation where, you 15

know, they have one nominal efficiency value for each 16

motor category, which is a combination of enclosure, 17

pole, and horsepower. 18

Now, if you have a bare shaft fan and you 19

need to calculate the wire to air, you know, what 20

enclosure do you choose? That's the first question we 21

had to find a solution to, and then what pole 22

configuration do you choose and what horsepower do you 23

choose? 24

And in the NODA, and again, like if you have 25

365


feedback on that, we can address each of these sort of 1

selection criteria for picking a default motor. In 2

the NODA, we didn't know what the pole would be. We 3

didn't have speed information. What we had was a lot 4

of market data on the pole configurations that are 5

sold on the market and some speed data from one 6

manufacturer. So we ended up using sort of a market 7

average across all pole configurations. The exact 8

weights are in the LCC spreadsheet and I can pull that 9

out afterwards. 10

Now, in selecting the horsepower for the 11

default motor, what we used was simply sort of a 12

sizing factor on the fan BHP which was of 1.2. And 13

that's pretty much it. 14

And then on the enclosure, since we wanted 15

to stay on the conservative side, we picked the 16

minimum nominal efficiency between enclosed and open, 17

whichever is the lowest. 18

Now let's say you're trying to evaluate a 19

fan that's sold with a TEAO motor, so you have your 20

electrical input power for that fan and you're going 21

to compare this to the maximum allowable electric 22

input power of a, you know, minimally compliant fan. 23

So we have that efficiency equation. We can get the 24

shaft input power. Now we need to get to the drive 25

366


system. And in order to have a more comparable, a 1

more fair comparison, instead of using the same 2

default values, we developed default values specific 3

to TEAO motors, which are currently not regulated. 4

And if we look at the data we found on the market, for 5

motors on the market, which are less efficient than 6

other regulated three-phase induction motors. 7

So what we did is we looked at, you know, a 8

list of catalog data from different manufacturers. We 9

looked at where those values, those nominal efficiency 10

values compared in comparison with the premium level 11

or the level of the regulation, and so that's what you 12

have on the left-hand side of that table. We use the 13

term NEMA Band to try to see sort of how many NEMA 14

Bands below NEMA premium are the TEAO motors, and that 15

sort of varied by pole and horsepower. And based on 16

that information we developed some TEAO default 17

values, which are kind of, you know, they're expressed 18

in terms of a number of NEMA Bands below the current 19

regulation. And then for the selection of the pole or 20

for the selection of the horsepower, then that would 21

be the same process. 22

And so that's just a summary. So, you know, 23

if you're trying to calculate the metric for just a 24

fan that's sold without a motor, without controls, 25

367


there's no motor, so you would use the default values 1

corresponding to the regulation for medium electric 2

motors. But that's only if it's a fan only. 3

Now, if it's a fan sold with a motor, or 4

later we'll see fans sold with a motor and controls, 5

if it's a regulated motor, you can just use the name 6

plate nominal efficiency of that regulated motor, and 7

for the calculation of the maximum allowable 8

electrical input power you'd use the default values. 9

So if your motor is better than, is performing above 10

the current regulation, the metric would show that. 11

MR. SMILEY: I have a question. Bill 12

Smiley, Trane, of course. So for a motor, you select 13

the motor based on 1.2 times the fan brake horsepower 14

to size or select what the appropriate motor 15

horsepower rating would be. But the fan will operate 16

on that motor at less than the motor full load. But 17

the default efficiency value that you are using is 18

based on the full load motor. So there would never be 19

a reason to actually test the motor. You would always 20

take the default. 21

MS. IYAMA: So there is actually another 22

component to the -- that's sort of how we pick the 23

default motor and its nominal efficiency and then in 24

the next slide we'll see how we calculate the part 25

368


load. 1

MR. SMILEY: Thank you. 2

MS. IYAMA: Now, if a fan is sold with a 3

motor that's not regulated, you know, you can't use 4

the name plate efficiency because it's not a regulated 5

motor. So instead, it's a TEAO motor, you could use 6

the TEAO default values that we just -- 7

(Audio echoing.) 8

MS. IYAMA: Hello? Okay. So that would be, 9

you know, when using the calculated based method to 10

establish the electrical input power of your fan. If 11

you're sold with a TEAO motor, you would use the TEAO 12

default values. 13

And then when you're trying to get to the 14

maximum allowable electrical input power, you know, 15

you could use the same default values. There's 16

different ways depending on what or how we want to do 17

this. You could also use the same default values. 18

But then, you know, you'd be sort of, you'd be 19

negatively impacted if you're selling your fan with a 20

TEAO motor because you'd be comparing your fan with a 21

TEAO motor to a fan with a non-TEAO motor. So that's 22

option one, option two. 23

And then for any other electric motor, you 24

know, for now we would consider them similar to how 25

369


bare shaft fans are rated, meaning there's no test 1

procedure for that motor if it's not possible to get 2

representative default values for these motors, rate 3

them as a bare shaft fan. So that's just a summary of 4

how we selected the horsepower -- 5

MR. WHITWELL: So sorry, could you explain 6

that again, what you do with the other electric 7

motors? 8

MS. IYAMA: So we didn't do anything. It's 9

just something that we need to think of. It's easy to 10

establish default values for regulated motors. It's 11

fairly easy to establish values for TEAO motors. But 12

there are other motors like, I don't know, split phase 13

or PSC motors, ECM motors that are out there that are 14

driving fans. And so for those, what should we do? 15

MR. HARTLEIN: Sanaee, one of the things I 16

don't see up there as well -- this is Dan Hartlein, 17

TCF -- is the possibility, if that's an unregulated 18

motor, or for any motor for that case, that a tested 19

value would supersede the default value. 20

MS. IYAMA: Yes. Right. 21

MR. HARTLEIN: So does that go without 22

saying I guess? 23

MS. IYAMA: Yeah. I think one thing to 24

highlight is that in that table it says calculation-25

370


based method, so that's just for that scenario where 1

we're -- 2

MR. HARTLEIN: Ah, okay. 3

MS. ARMSTRONG: Hold on. What do you mean 4

by tested value? 5

MR. HARTLEIN: Meaning that if I take a -- 6

MS. ARMSTRONG: Tested value for the whole 7

system or just the motor? 8

MR. HARTLEIN: If I take a motor driven fan 9

and I measure the power in and the fan performance 10

out -- 11

MS. ARMSTRONG: I got it. That's fine. 12

MR. HARTLEIN: -- that should substitute. 13

MS. ARMSTRONG: Absolutely. So the 14

question, though, is do you want that always to trump 15

or do you want the manufacturers still to have the 16

option? We can set the regs up either way. 17

MR. HARTLEIN: I'm not sure. 18

MS. ARMSTRONG: Well, think about that. 19

That's something we'd like feedback on obviously. 20

What's not here -- 21

MR. HARTLEIN: I have a lot of votes for 22

option, but I'm not sure. Let me think about it. 23

(Laughter.) 24

MS. ARMSTRONG: What's not here is testing a 25

371


motor. You either test the system or you use the 1

default values. So that's where we are. And like 2

Sanaee said, we don't have default values for all 3

types of motors, so if we end up finalizing as 4

proposed and you have a different type of motor that's 5

not in here, you would need to test the system, the 6

full wire to air, yeah. 7

MR. HARTLEIN: And I'm assuming we could 8

extend this to include controls like a VFD as well, in 9

the same discussion basically. 10

MS. IYAMA: So we'll get to that in the next 11

slides. 12

MR. HARTLEIN: Okay. 13

MR. SMITH: This is Wade. I'm not 14

challenging what you said at all because I'm not sure 15

what position we would want to take in any event, but 16

is there a reason why you wouldn't allow a 17

manufacturer to characterize a particular motor and 18

then use it on several different fans with 19

mathematics? Test a particular motor, characterize 20

it, and then use the results of that test coupled with 21

fan tests? 22

MS. ARMSTRONG: So I think from a high level 23

point of view, in theory, I don't think or I guess in 24

principle I don't think DOE has an issue with 25

372


something like that. What we don't have is test 1

procedures for all different types of motors to make 2

sure they're tested in an equitable manner since all 3

the defaults are derived in an equitable manner. So 4

that's what's missing, and we wouldn't want to be a 5

manufacturer-specific declaration of a method of test 6

to get a motor efficiency to then use as default 7

values. So, like I said, we're not necessarily 8

opposed to the idea, but the hurdle is the absence of 9

a test procedure that exists for a lot of other kinds 10

of motors that are currently not subject to standards. 11

MR. FLY: This is Mark Fly with AAON. 12

Ashley, and I'm not familiar at all with the motor 13

test standards, but I'm sure you are, but are they 14

looking at part load efficiencies in part of that 15

standard, or is it a full load only test? 16

MS. ARMSTRONG: Right now it's full load. 17

MR. FLY: So your test method is only full 18

load. 19

MS. ARMSTRONG: Right now. 20

MR. WHITWELL: Ashley, this is Bob from 21

Carrier. So a follow-up question on this, and maybe 22

I'm way ahead of where we need to be on this. So, if 23

we're using regulated motors and we have multiple 24

regulated motors that can go into a particular 25

373


product, can we test the smallest or let's say our 1

base regulated motor knowing that higher horsepower 2

motors are going to be higher efficiency? 3

MS. ARMSTRONG: I think that's something for 4

the group to discuss. You know, these fans that are 5

coming up with the operating type metrics that operate 6

in a certain range don't, for lack of a better term, 7

cross-pollinate parts. They would be rating every 8

point for that specific fan and system in the testable 9

configuration. 10

So what you guys do for air conditioners and 11

what DOE allows you to do is this concept of you can 12

test a base model and then you can switch out 13

otherwise larger types of motors and not have to test 14

and rate all of them such that that motor's efficiency 15

is roughly equivalent to or more efficient than the 16

base motor is the concept in theory. 17

There's nothing like that in fans right now 18

at least that we're considering. I guess if we had a 19

reason to consider it or you had a suggestion or 20

alternative that you wanted us to consider we could 21

talk about it. 22

MR. WHITWELL: Yes. So I guess as we go 23

forward we need to think about those kind of things, 24

because I'm concerned about the test burden that we'll 25

374


be under. 1

MS. ARMSTRONG: Why wouldn't you just use 2

the defaults? 3

MR. WHITWELL: I don't know. I don't know 4

yet. 5

MS. ARMSTRONG: I mean, that's what the 6

defaults get you, right? 7

MR. WHITWELL: Maybe. Maybe we will want to 8

use the default. I don't know. 9

MS. ARMSTRONG: To the extent you get a 10

different default value because your motor has 11

different properties, but that's what the defaults get 12

you. They get you around that test burden in all 13

those different combinations in theory. They give you 14

default values so it's just a calculation. 15

MR. WHITWELL: So, okay, so we could -- 16

yeah, all right, I see. We use the default value for 17

all the different motor combinations. 18

MS. ARMSTRONG: You could. I mean, unless 19

you wanted to run the full test, you just use the 20

default and run calculations. It's kind of like, in 21

essence, it's partially an AEDM that we're putting in 22

regs, a standardized one. 23

MR. SMILEY: Bill Smiley, Trane. Along 24

those same lines, Bob, I think you would have to 25

375


somehow in your test measure directly the fan power in 1

order to use the default, because you need the fan 2

power in order to apply the default of the motor. So 3

you'd have to do the test and actually measure the fan 4

power. 5

MS. IYAMA: So I want to go back to this 6

slide because I think it's relevant to what Bill's 7

saying. So, in a calculation-based method, which is 8

when we would use the default values, this is where 9

you would measure. It's the shaft input power. And 10

in the wire-to-air test you wouldn't be using -- it's 11

that bottom row of the table. The test output would 12

cover the whole thing, so you're not using default 13

values. 14

MS. ARMSTRONG: Right. So, to answer your 15

question directly, if you use option one, you would 16

use these defaults for your different combinations, as 17

long as you got to the same default you would get to 18

the same default rate. 19

For the bottom one, though, the bottom ones 20

would show up with different variations to the extent 21

the motor transmission controls the result and 22

variances that affect the FEP. But your default would 23

be a way to streamline. So we've implemented here a 24

different way to get to the same I think burden 25

376


reducing type of scheme. 1

MS. IYAMA: Okay. So that was for the 2

default values. Just a summary and we could also get 3

feedback on those points that would be useful. So the 4

sizing of the motor based on the 1.2, we could do it, 5

you know, with a different multiplier. We could do it 6

based on constant motor horsepower, especially when 7

the fan is sold with a motor. For the motor enclosure 8

we chose the enclosure leading to the lowest nominal 9

full load efficiency. We could also do it as a sales 10

weighted average like we're currently doing for poles. 11

These are things that are up for discussion. 12

And then the next slide here is the equation 13

that was used to get to part load efficiency. And, 14

you know, we're also reviewing the draft 207 that AMCA 15

put out. They also have another way of doing this. 16

There's also, you know, in Europe other ways of doing 17

this. This is how they did it in the pumps 18

rulemaking. So the first equation here is just the 19

losses of the motor at part load equal the full load 20

losses times a factor, and that factor is a function 21

of the load. And where did we get those coefficients? 22

Well, we got them from analyzing data that was 23

submitted by NEMA during the pump rulemaking process 24

and ASRAC process. 25

377


MR. SMILEY: Question. Bill Smiley, Trane. 1

So that polynomial curve fit was supplied by some 2

data that you got from NEMA and is that -- that's 3

assuming that every motor, every size has an off 4

design efficiency reduction of the same amount? 5

MS. IYAMA: Yeah, and in a very conservative 6

way. It's a pretty -- 7

MR. SMILEY: Do you have the data that NEMA 8

gave you? Is it a function of horsepower and poles or 9

what? 10

MS. IYAMA: I think it's probably published 11

as part of the pumps rulemaking. It's probably in the 12

docket there. I don't have it with me. 13

MS. ARMSTRONG: How about we'll check and if 14

it's in there we'll circulate it. We can also get 15

someone from NEMA Motor Coalition to come in here at 16

the next meeting and kind of present what they did, 17

because they were the driving force behind this 18

estimate in the pumps negotiation. 19

MR. SMILEY: I think this is excellent. My 20

interest is -- 21

MS. ARMSTRONG: Are they the same type of 22

motors. 23

MS. IYAMA: I can tell you about the 24

methodology. 25

378


MR. SMILEY: I would like to see what the 1

data is because I could use that in some unit design 2

work that I do. 3

(Laughter.) 4

MR. SMILEY: Okay? If it's available. 5

Thank you. 6

MR. FLY: This is Mark Fly with AAON. As 7

you went through this and you did the 1.2 times the 8

fan, as I sell this in a piece of my equipment, if I'm 9

not 1.2 times the fan horsepower above, do I rate it 10

at 1.2 or do I rate it at the actual applied? Because 11

maybe I'm 1.0 because that's what the customer -- I 12

mean, basically we let the customer select the motor 13

and the fan, and what I'm going to have to do with all 14

of this as I'm thinking ahead is I have to be the 15

application police to make sure that they select 16

something within that's legal. 17

MS. IYAMA: There's the case when you're not 18

selling a motor with your fan, and then you need to 19

know what's your multiplier to select the default 20

motor and you want everybody else to do the same, so 21

that would be 1.2. And then the case when you're 22

selling a motor with your fan. And then, you know, do 23

you use the actual horsepower of the motor provided 24

with your fan? Keep in mind that you would be 25

379


comparing that electrical input power to the 1

theoretical one, the maximum allowable one, and for 2

that one the motor would probably be sized, you know, 3

with the 1.2 factor or whatever factor you agree on. 4

MR. SMILEY: Well, I think what you're 5

saying -- Bill Smiley, Trane -- is that if you test 6

with a motor, you don't need to worry about the 1.2 7

because you're testing with the motor and the motor 8

sized at .9 or 1.4 or whatever. 9

MS. IYAMA: Right. 10

MR. SMILEY: Because a lot of times it 11

depends on the load compared to the motor full load, 12

where 1.2 may not be -- 13

MS. IYAMA: Right. 14

MR. SMILEY: But if you don't know any of 15

that, you use the 1.2 in order to develop into the 16

default values, right? 17

MS. IYAMA: That's right. Correct. 18

MR. SMILEY: Correct. Thank you. 19

MR. FLY: This is Mark Fly with AAON. What 20

I want to do is click the box that says this is a good 21

fan in this unit, right? So do I need to analyze it 22

with the motor I supply? If I'm supplying a motor, is 23

that always the case, or has that not yet been 24

decided? 25

380


MS. ARMSTRONG: So what I heard I think at 1

the beginning of it is you guys want the option to 2

elect to do -- potentially. We're still deciding, 3

but, you know, that you guys want the option to do one 4

way or the other at your discretion. And so we either 5

standardize it with the 1.2 nominal or you do the full 6

test. In other words -- 7

MR. FLY: Because one way is saying at this 8

application point this fan alone is valid. 9

MS. ARMSTRONG: Yep. 10

MR. FLY: You know, with all the assumptions 11

attached. And the other way is saying that at this 12

particular operation point this assembly with the 13

controls and everything through a calculation default 14

method is valid. So I think it could give you two 15

different answers. 16

MR. DIKEMAN: Ashley, let's go this other 17

way. Above and below the line. 18

MS. ARMSTRONG: Yeah, I hear what we're 19

doing. I think we were -- 20

MR. DIKEMAN: No, no, no, please. Here's 21

the motor losses. If they're the same above and below 22

the line, the default VFD losses are the same above 23

and below the line, it all comes back to impeller BHP. 24

That's the difference between the required 25

381


performance and what you're actually doing. Default 1

is the same above and below the line. It washes out. 2

It comes back to impeller BHP. 3

Now, in Dan's case, he brings in a different 4

case. He tests this motor drive bumper to bumper. 5

Now he can bring a new performance into the 6

conversation. But defaults are above and below the 7

line, they wash. You don't have to be motor -- you 8

used a six pole motor, but the rulemaking is done on 9

an average motor. You're not getting penalized for 10

that either unless you test the six pole motor. Now 11

you're dragging along some additional inefficiency. 12

MR. HARTLEIN: This is Dan from TCF. I 13

would also add to that that there are certain areas, 14

for example, high-performance axial fans where we 15

probably know more about that air over motors 16

performance with that flow than NEMA does and we're 17

better at it than they are, so we can do things with 18

that fan and design that the standards that NEMA would 19

provide or the motor manufacturer wouldn't have 20

because we've actually tested that fan in that 21

configuration, and so therefore we know actually what 22

that motor is consuming because of its air over and 23

the capacity for us to get additional cooling to that 24

motor and keep them at optimum temperature. So we, in 25

382


some areas we've actually gone beyond where the motor 1

guys are in applying their product and ours. 2

MS. IYAMA: Dan, just on that, this equation 3

was developed based on data provided by NEMA, but the 4

default TEAO values that I showed, that was not. That 5

was just us looking at catalog data, TEAO ratings that 6

are available. 7

MR. HARTLEIN: Yeah. Thank you, and I would 8

just like to add that I didn't state that we knew more 9

than the DOE. 10

(Laughter.) 11

MS. IYAMA: All right. So next slide is -- 12

MR. SMILEY: Excuse me, Bill Smiley, Trane. 13

This equation does not apply to TEAO or it applies to 14

any motor? 15

MS. IYAMA: It would apply to, yeah, all 16

types. Yeah. 17


MS. IYAMA: So here I was going to put the 19

equation we use for the transmission efficiency, but 20

those slides were actually not really finalized, but 21

they're in the spreadsheet, so we can go over that a 22

little later. But the main thing is it's the same 23

form then, what AMCA uses in its AMCA 203 standard, 24

and I don't know how familiar you are with that 25

383


document, but in that document they have three 1

equations to represent belt efficiency, one with 2

higher losses, one with medium losses, and one with 3

low losses, and we use the high losses because we want 4

to be conservative. 5

Now for fans with motors and controls -- 6

MR. WOLF: Sanaee, Mike Wolf here, at the 7

expense of slowing you down here a little bit. Mike 8

Wolf with Greenheck. So, with the transmission 9

losses, you used a worst case even though there's 10

three different scenarios in AMCA 203, right, I think? 11

And you said you did it because you wanted to make a 12

worst-case conservative rating I guess, if you will. 13

Why would we not have used that same 14

methodology or assumption with all the motor data that 15

we ran? And the reason I ask that is I'm starting to 16

think ahead a little bit to what we're going to 17

publish to the field and how this whole thing is going 18

to work down the road, you know, kind of relative to 19

the catalog page Tim showed here earlier where you've 20

got the products that are in or the selection range in 21

compliance and then those that are out. And I'm just 22

starting to get concerned that okay, so we're going to 23

have different motors that might be in compliance or 24

out of compliance depending on what motor default 25

384


coefficient we use for our rating. 1

Steve, you're looking puzzled. I'm not sure 2

I'm being clear. So I guess to simplify the question, 3

why don't we just the worst case for the motors too 4

and keep it simpler? 5

MS. ARMSTRONG: I agree. 6

MS. IYAMA: So here, when we say that we're 7

using the nominal efficiency values as established by 8

the regulation, that's pretty conservative. 9

MR. WOLF: Right, but go to your slide where 10

you have the green. Yeah, okay. So there. 11

MS. IYAMA: So these are going to be the 12

minimal nominal efficiency values that you're going to 13

be finding on the market once the new medium electric 14

motor regulation comes into force. 15

MR. WOLF: Okay. So it's a minimum based on 16

the horsepower. But then you had another one for the 17

TEAO I thought that you came up with. 18

MS. IYAMA: TEAOs are even lower because 19

these are nonregulated motors, and we felt like since 20

they're so -- they represent a pretty significant 21

share of the motors that are sold with fans, we should 22

try to develop default values for those, especially 23

because they're even lower than the ones on the 24

market. 25

385


MR. WOLF: Just to make sure I'm 1

understanding this right. So going forward if we're 2

going to look at say a given fan, let's say running at 3

one, roughly one horsepower, it might have two 4

different ratings, two different, and I think I'm 5

saying this right, two different FER calculations 6

depending on whether I'm calculating the FER with a 7

totally enclosed air over or a standard motor or if 8

I -- 9

MR. DIKEMAN: Or a belt drive. 10

MR. WOLF: Say that again, Steve. 11

MR. DIKEMAN: Steve Dikeman. Or a belt 12

drive. 13

MR. WOLF: Well, but I'm assuming they both 14

have belts. 15

MR. DIKEMAN: Okay. 16

MR. WOLF: Okay. But that would be a fourth 17

one. You've got a direct drive without belts, and 18

then you could have another one where you've got your 19

own motor that you've done wire to air, so I'm just 20

trying to think through how many permutations of 21

ratings are going to be out there. 22

MS. ARMSTRONG: So, as you presented the 23

question that way, I think you raise a completely 24

valid issue and that is if a given fan is offered with 25

386


a number of different motors, which default value do 1

you choose? And do we really need -- does the 2

committee think we really need a variety of defaults, 3

or can we just say this is the default motor? It's 4

not by type. Do we need to actually characterize it 5

by type, by motor type? Or can we just say this is 6

the least efficient motor and if you use default 7

values, that's what you get? 8

MR. WOLF: So Mike Wolf here again. So I'm 9

thinking about, okay, we're going to establish some 10

base level here, and ultimately what we're trying to 11

do is drive the manufacturers and the market to more 12

efficient, better selections. So, as I go through the 13

thought process here with looking at how I calculate 14

an FER, well, one way to get a better FER is to 15

eliminate belts, right? Because I have a worst-case 16

scenario, so I think that's obvious. I'm just 17

wondering if we've got multiple motors, you know, 18

we're going to drive people to a certain motor, which 19

that's probably the desired outcome here. I just want 20

to make sure that what we're doing is driving the 21

right behavior after this is in place. And having 22

multiple metrics I don't think is the important thing. 23

The important thing is having the right things 24

accounted for so that we can again -- 25

387


MS. ARMSTRONG: So I think a question, 1

though, is do you guys want the ability to rate 2

different kinds of motors, perhaps differentiate 3

between more efficient kinds of motors in a different 4

way other than tests? Because if we don't provide 5

default values for different kinds of motors you 6

preclude yourself from ever differentiating by motor 7

efficiency if you don't test full wire to air. So 8

that's really a question for you guys. 9

MS. IYAMA: I can give another example just 10

to illustrate the sort of issues that you would think 11

through. Just here, for example, so when you're using 12

a calculation-based method and you're trying to 13

establish the electrical input power of your fan, but 14

you only did a bare shaft test. So, if you were to 15

use the TEAO default values to represent the motor 16

that you're selling with your fan or if you were to do 17

a testing-based method where you would, you know, 18

establish electrical input power of your fan based on 19

the TEAO performance, you would be comparing that to 20

the column that's right on the left here where the 21

electrical input power of your minimally compliant fan 22

is calculated with the default values. If that 23

default value is equal to the regulation, you may be, 24

you know, in a different position than if that default 25

388


value is equal to those TEAO values that we just 1

showed that are lower. 2

MR. MATHSON: A comment. Tim Mathson from 3

Greenheck. A couple of comments. One is I think what 4

we're weighing here is simplicity with accuracy 5

probably. That's kind of what it sounds like. If we 6

go with single numbers for defaults, that makes it 7

more simple but maybe not quite as accurate. So that 8

may be something that we're weighing. 9

Another comment that I would say is the AMCA 10

203 belt losses or drive efficiencies that you get 11

from AMCA 203, in AMCA 207, which is still under 12

development, we're using the middle curve, that 13

average one. And, you know, nobody has good 14

information on belt drive performance. I only have 15

the information, I've tested maybe a dozen 16

combinations and never gotten as bad as that most 17

conservative line. So I personally think it's a 18

conservative standard, but again, that's only on a 19

dozen drive combinations. 20

And the third thing to keep in mind is that 21

as we compare belt drive with direct drive, we want to 22

incentivize the use of direct drives, but we want to 23

be somewhat realistic because a belt-driven fan 24

running at 800 RPM still uses a four pole motor, which 25

389


is much more efficient than an eight pole motor. So 1

it may be weighted a little bit too much that way if 2

we use a single number for all the poles. Does that 3

make sense? 4

MR. PERSFUL: This is Trinity with Clarage. 5

To address Ashley's question about should you use one 6

default value regardless of type of motors or 7

multiple, I would say it depends on what type of 8

behavior we want to drive. If you use just one, I 9

think in my mind I call that the least common 10

denominator, and it's going to drive -- and if it 11

happens to be the cheaper motor, it's going to drive 12

people to use the cheap, less efficient option. If 13

you have multiple ones, it's now going to reward 14

people for using a more efficient motor. So I would 15

suggest that you look at it. 16

MR. WOLF: I think we need to think about 17

this and get back to it. 18

MS. ARMSTRONG: Well, so if we pick a 19

default motor and it's just one nominal value and that 20

one nominal value is the same value used across the 21

board no matter what, you would never be able to claim 22

the savings you might get from using a more efficient, 23

different type of motor, thus incentivizing -- you 24

would lose that differentiation, period, unless you 25

390


test. 1

So I think to Mike's point, if you want to 2

incentivize the differentiation and you're going for 3

labeling differentiation, utility program 4

differentiation, et cetera, et cetera, what you really 5

want is default values, conservative default values, 6

but default values for all the different types of 7

motors, that we can come up with reasonable default 8

values that they're tested the same. At least then 9

you wouldn't necessarily see the difference in 10

efficiency within a motor category, but you would be 11

able to see that certain types of motors, perhaps 12

those that are subject to standards, are more 13

efficient just generally than other types of motors 14

that are currently unregulated, like TEAO motors, and 15

you may be able to make better choices. 16

MR. DIKEMAN: So, Ashley, may I take that 17

one step further? If the default motor efficiency is 18

reflective of four pole, just for purposes of 19

conversation, and you happen to live in a 1200 six 20

pole motor world, and you went and tested your six 21

pole motor, you're going to struggle to get back to 22

the bench line. So, if 1200 can stand by itself and 23

1800 can stand by itself, I think you'll get more of 24

what you want. 25

391


MR. HARTLEIN: So I would argue a little bit 1

to the contrary. Dan from TCF. And I think you have 2

to look at the behavior of fan manufacturers to 3

totally understand this, and if there's an advantage 4

to be gained through a testing program, it's going to 5

be done. So, if I can save a motor class, if I can 6

save a motor frame, drop to a lower frame size because 7

I went through the testing, attached a VFD, did all 8

that work, I'm going to do it because it gives me an 9

advantage in the marketplace. I'm going to rate my 10

product according to that. 11

If I don't need those default values to 12

incentivize the operation to do that, because we're 13

going to do it because we're going to take that 14

advantage. Because if I can stay down a frame size in 15

motor and my competition can't, I'm going to get the 16

order and they're not. So I would think it's going to 17

happen anyway. That's basically the way we're going 18

to behave. 19

MS. ARMSTRONG: Okay. So you guys might 20

want to think about this a little bit. I think we can 21

do it either way. Obviously DOE, as a policy 22

position, we want to incentivize more efficient 23

systems, so we want to give you the tools, whatever 24

that is necessary. 25

392


One thing to think about when you're talking 1

about your method, Dan, is that, I mean, you're 2

talking about testing every variation out there. If 3

you have a fan that an accessible configuration is 4

offered with different motors, is offered with -- you 5

would then be testing -- 6

MR. HARTLEIN: Can I ask a question? This 7

is Dan, to that. 8

MS. ARMSTRONG: Yes. 9

MR. HARTLEIN: Not having been through the 10

gristmill of DOE regulation before, so if I as a 11

manufacturer take the information and knowledge that I 12

have on how my products perform across the range with 13

particular motors and things and I choose to make the 14

decision to rate that at a certain point, so I apply 15

those standards to rate that product and to represent 16

it as being something that I'm sure it is, then I'm 17

falling well within the reg. I don't have to test, 18

right? The only thing is you may test and I better be 19

right. Is that correct? 20

MS. ARMSTRONG: No. 21

MR. HARTLEIN: No, it's not? 22

MS. ARMSTRONG: No. 23

MR. HARTLEIN: Okay. So help me understand. 24

MS. ARMSTRONG: So the way this works is 25

393


your ratings must be based upon the DOE test procedure, 1

must, and applicable sampling provisions. So, if we 2

don't provide you with a method in the test procedure 3

to either extrapolate to other ratings, nontested 4

ratings, so calculation-based ratings, you must test 5

every single model that you're going to rate. And when 6

you would submit that information to DOE, you would 7

sign a legal binding statement that says I have 8

developed these ratings in accordance with the DOE test 9

procedure and sampling provisions. They're tried, 10

true, accurate, et cetera, et cetera. So no, you can't 11

just rate without testing. 12

MR. HARTLEIN: So therefore -- this is Dan 13

again -- a custom fan has to be tested. 14

MS. ARMSTRONG: So not necessarily. 15

MR. HARTLEIN: Okay. So if the rule allows. 16

MS. ARMSTRONG: We have provisions for 17

dealing with that stuff, but they need to be provisions 18

in this rule. So the moral of the story is that if you 19

want to have what you call untested ratings, whether 20

that be calculation-based ratings, whatever, we need to 21

make sure that the regulations allow for that. 22

MR. HARTLEIN: Okay. Gotcha. 23

MR. FLY: This is Mark Fly. Dan, I need to 24

talk to you about your motor supplier because if you've 25

394


got a motor supplier that's never going to do a design 1

change and that you can buy from the same guy every 2

day, I need to know who that is, because otherwise 3

you're going to have to retest when that motor supplier 4

changes their efficiency or you have to -- I can't get 5

this one this week. I have to go buy somebody else's. 6

Or you have to test all of them and rate on the 7

minimum one. 8

MR. HAUER: Ashley, it's Armin Hauer 9

speaking. I have to jump ahead and ask about labeling 10

and certification. Is it sufficient that I put on a 11

label compliant, yes/no? Or do I have to put on 12

percentage 92.8 percent? That's really a difference 13

because it's very easy for if you have your process 14

under control as a motor manufacturer to basically just 15

stick your head out the window and say yes, my fan is 16

going to be compliant, I don't have to test because I 17

have the experience. 18

MS. ARMSTRONG: Okay. So I think the answer 19

is I'm going to split the parts. 20

Certification is different than labeling for 21

DOE. We can definitely talk about what you guys want 22

your labels to look like, whether it's a yes/no type of 23

scheme or it's an actual number type scheme. What that 24

number is, what the metric is, we can talk about those 25

395


types of things. Certification, you must submit 1

paperwork to DOE that shows your products are in 2

compliance or self-certifies that your products are in 3

compliance before distribution and commerce in the U.S. 4

That includes importation. 5

So that being said, our certification scheme 6

for all the places where we have efficiency or 7

consumption metrics right now, you tell us the number 8

for every single product that you have and that number 9

must be based on the DOE test procedure and sampling 10

plan. Now many of our commercial equipment allows for 11

estimations based on modeling or calculation-based 12

methods as a burden reducing measure because we 13

understand there are custom products. Not everyone has 14

the ability to test everything, et cetera, et cetera. 15

So that's all open, but it all has to be provided by in 16

regulation. So I would assume that at the end of the 17

day you will have to certify the actual efficiency 18

values and then the content of the label is up to you. 19

MR. FLY: And this is Mark Fly with AAON. 20

And so as part of this we're going to have to define 21

models, which is one of my favorite subjects, and so 22

you're going to list a model with the DOE -- 23

MS. ARMSTRONG: So I think it's a little 24

easier for this one because we're doing this range 25

396


concept. But you're right. We do need to talk about 1

if you have a given fan, and this is what I don't have 2

a true appreciation of in your industry, but if you 3

have a given fan, what are all the variations thereof 4

that get added to that fan, that same I guess -- I 5

don't know the best way to say this. I mean, some may 6

be non-efficiency, some may be efficiency related. 7

That's kind of a conversation we need to have at some 8

point because we need to translate that into how many 9

different ratings you will end up having to tell the 10

department. That's kind of how the regs work. 11

But I think that's one of the reasons you 12

see this scheme from the department. We were trying to 13

give you options for not having to test everything. 14

This is a calculation-based method, which is different 15

than a full-blown AEDM. I still call this based on 16

testing because you're still testing the fan component, 17

but it allows you to calculate the downstreams. So you 18

guys think about it. That's kind of how in high level 19

terms our regs work. Yeah, go ahead. 20

MR. SMILEY: Bill Smiley, Trane. If you go 21

back to your slide where you had the motor poles and 22

efficiency for the DOE regulated ones, yeah, that 23

chart, now you look at the TEAO, which is the next 24

chart, and you have efficiency values as a function of 25

397


motor speed or number of poles. But if you back up a 1

slide, you say oh, okay, for all these motors we're 2

just going to have one number that represents 3

everything all the way across there. Is there a reason 4

for that? I mean, we could actually do the same thing 5

to this that we did to TEAO. 6

MS. IYAMA: They're used for two purposes, 7

and I think that's where the difference is. If you're 8

selling or if you have just a fan without a motor, you 9

don't know what the pole is. You can't, if I give you 10

this table, you don't know what value to pick, the two 11

pole, the four pole, six pole, eight poles, 12

MR. SMILEY: But isn't it the same with 13

TEAO? 14

MS. IYAMA: If a fan's sold with a TEAO 15

motor, if now you're trying to establish the electrical 16

input power of a fan sold with a TEAO motor, you know 17

what the pole of that motor is. 18

MR. SMILEY: Not necessarily. It could be 19

belt drive. 20

MR. WHITWELL: Or it could be the fan. Bob 21

Whitwell. 22

MS. IYAMA: But if the motor is provided 23

with the fan. 24

MR. WHITWELL: Okay. But there are other 25

398


cases where we buy fans and we might apply them with a 1

TEAO motor in some case or we apply them with a 2

regulated motor in other cases. 3

MS. ARMSTRONG: Right. So what Sanaee has 4

set up here originally, it's for discussion purposes. 5

If you just have a fan, you're actually getting a 6

better number than you would as if you had a fan and 7

you know you're ultimately going to end up in a TEAO 8

setting, but you don't know the specifics of that 9

motor. Because if you look at the two numbers -- I 10

mean, you're getting some benefit there that maybe 11

you're arguing you shouldn't be getting. 12

MR. WHITWELL: Well, no, no. My argument -- 13

I'm with -- 14

MS. ARMSTRONG: We can go back. 15

MR. WHITWELL: I think I'm with Bill that on 16

the -- if we go back to the regulated motors -- 17

MS. ARMSTRONG: Yeah, except for what she's 18

saying is if you have a fan, irrespective of motor, 19

that even if you know the motor it's going into, but 20

you don't know all the motor's characteristics, you go 21

to that first table. So you know it's air over, but 22

that's irrelevant to the way we set this up. So you're 23

actually arguing for us to set up default values for 24

I'm a manufacturer -- 25

399


MR. WHITWELL: Sorry. Forget the air over 1

part of it because that's a small piece. I'm more on 2

the -- 3

MS. ARMSTRONG: So just go to one. 4

MS. IYAMA: Okay. So okay. So I guess 5

here, let's go to this. 6

MS. ARMSTRONG: I have a fan, right? 7

MS. IYAMA: The first row. I have a fan 8

only, no motor, and I don't know anything about that 9

motor that's going to be -- 10

MS. ARMSTRONG: You don't know any 11

characteristics of that motor. You may know the 12

specific category, but that's it. 13

MR. WHITWELL: Right. So then I go to 14

the -- so, okay, so what's the basic default then? I 15

guess I don't understand then. Is it the regulated 16

motor default or is it TEAO default? Where do I go? 17

MS. IYAMA: So I'm going to try to explain 18

how to read that table and then maybe it's going to 19

clarify. 20

MR. WHITWELL: Thank you. Sorry, I'm slow 21

about this. 22

MS. ARMSTRONG: It's okay. 23

MS. IYAMA: If it's a fan being sold without 24

a motor, without transmissions, without anything. It's 25

400


just the non-driven fan, bare shaft fan. When you're 1

calculating the electrical input power based on the 2

calculation-based method, you're going to use default 3

values for the motor which are the default ones. The 4

default ones are equal to the minimum efficiency 5

standard that is in place for medium electric motors. 6

If your fan sold with a motor and that motor 7

is a regulated motor, you can use the name plate 8

efficiency of that motor -- 9

MR. WHITWELL: Without testing. 10

MS. IYAMA: -- as your default value in the 11

calculation-based method when calculating -- 12

MS. ARMSTRONG: So right now this is all up 13

for discussion, right? This was our going-in kind of 14

how this might all work. But obviously let it sink in. 15

We'll red line this to reflect some of our discussions 16

maybe not to send out tonight, but give us a day or so. 17

MR. SMILEY: Well, my question originally 18

was really, didn't have much to do with any of that. 19

It was why do we have efficiencies as a function of 20

poles on one type of motor and not as a function of 21

poles on the other type of motor -- 22

MS. IYAMA: I can answer that. 23

MR. SMILEY: -- because if we have to 24

develop a process to take a default value, it's a table 25

401


look-up. 1

MS. IYAMA: I'll try to answer that. 2

MR. SMILEY: But you're actually right, 3

though, if we don't know what the motor poles are, then 4

we have to have a value to use, and that could also be 5

the situation for TEAO. 6


MR. SMILEY: So which one of those -- we can 8

use whichever one we want or -- 9

MS. ARMSTRONG: So we're going to clarify 10

this table first of all because I think that some 11

clarification of this table would go a long way. 12

MR. SMILEY: That was really -- it's kind 13

of -- I agree with everything you're doing. 14

MS. IYAMA: I should have introduced it in a 15

better way, and I'll try to clarify, but it's really 16

two tables in one where you have a situation for when 17

you're just a fan without a motor and you don't know 18

anything about the fan, and in that case, how do you 19

calculate your wire to air while you use those default 20

values. And you don't need to worry about which pole 21

to pick. It's just one value by horsepower. 22

MR. SMILEY: Or what type of motor. 23

MS. IYAMA: Or what type of motor. Exactly. 24

And then there's a different situation, 25

402


different situation for a fan sold with a motor where 1

you know what the main characteristics of your motor 2

are, so you can identify if you're in a case where 3

you're a regulated motor, a TEAO motor or another kind 4

of motor. And based on which case you're in, you're 5

going to either use the name plate efficiency of your 6

motor, the TEAO efficiency of the pole and horsepower 7

that corresponds to the motor supplied with your fan, 8

and for other motor we don't know yet what we want to 9

do. 10

MR. SMILEY: So basically you're saying on a 11

TEAO motor there's no name plate efficiency, so you 12

have to define what that is. 13

MS. ARMSTRONG: Yes. 14

MR. SMILEY: Is that what you're saying? 15

MS. IYAMA: Yes. 16

MS. ARMSTRONG: They're not regulated. 17

MR. SMILEY: Okay, thank you. That would 18

have been the answer to my question. 19

MR. WHITWELL: Thank you. 20

MS. IYAMA: Sorry. 21

MR. AUTH: Chris Auth, Baltimore Aircoil. 22

The TEAO motors we use are name plated by NEMA. I 23

don't know if they're regulated or not, but there is a 24

nominal efficiency on the name plate. 25

403


MS. ARMSTRONG: It's not required, so some 1

do it and some aren't. So we can't depend on it being 2

there. It's not -- 3

MALE VOICE: It's not a regulated value. 4

MR. AUTH: But would that name plate value 5

work in this analysis? 6

MS. ARMSTRONG: No. I mean, you can argue 7

that it should, but not everyone's doing it the same 8

way, as we have learned through experience. 9

MR. SMILEY: If that's not regulated, 10

there's nobody saying that that's the true number. 11

MR. WHITWELL: My understanding for most of 12

the motors that we purchase, they're basically, they're 13

actually TEFCs with the fan removed, but these are 14

larger motors. It's not always the case, but I'm 15

just -- I believe the motor is tested. 16

MS. ARMSTRONG: So TEFC, whatever, those are 17

regulated, so that's why. Those are in a different -- 18

MALE VOICE: (Away from microphone.) 19

MS. ARMSTRONG: Right, but that's why you're 20

seeing name plates. They're required. 21

MR. WHITWELL: So that rating's not right 22

because the cooling fan internally -- 23

MS. ARMSTRONG: So you want me to call NEMA 24

tomorrow? 25

404


MR. SMILEY: No. I think what you're doing 1

is -- 2

MS. ARMSTRONG: Okay, sufficient. 3

So a couple things. I do want to wrap up 4

this discussion in about 10 minutes or so. We're about 5

30 minutes out from our end and I do want to do some 6

recaps of where we are, where we're going, kind of 7

homework items, what we've promised around the table 8

before do actually wrap up. 9

MR. BUBLITZ: Mark Bublitz, New York Blower. 10

Sanaee, I want to circle back to nominal efficiency 11

values for motors. If all my components are right at 12

the margin and I acquire a regulated motor and that 13

motor performs under the nominal value, but it's still 14

regulated, I would fail the test because that minimum 15

nominal is not a minimum value. So I just want to 16

throw that out there as what's the true floor of that 17

test. 18

MS. ARMSTRONG: So this came up and I think 19

I'm going to say your question a different way. But I 20

see that as, DOE, how would you enforce. So, in other 21

words, DOE, if you're going to enforce my rating and I 22

had developed it through the application of the nominal 23

values, recognizing that those nominal values are 24

really closer to average values, not the least 25

405


efficient of the population and I happen to get the 1

least, DOE, if you tested that and it came out below, 2

if you did the full wire-to-air test and it came out 3

below, what would you do? I mean, that's how I 4

translate what you're asking me. 5

MR. BUBLITZ: That's fine. I'm going to say 6

it's accurate. 7

MS. ARMSTRONG: I think that's up for 8

discussion. One of the things that we have discussed 9

is DOE enforcing the same way you generate your rating. 10

So, if you do it with nominal values, we would do it 11

too. We would test your fan and we may get different 12

fan numbers, but if you applied nominal values, we 13

would too. 14

Alternatively, we could all do testing 15

values representing, with the realization that those 16

nominal values should be conservative and testing 17

should always result in better or greater. Really, 18

it's up for discussion. 19

So do you guys want to keep going? How many 20

more slides do you have? 21

MS. IYAMA: I have two on the controls and 22

then three on the test procedure. 23

MS. ARMSTRONG: So do you want to talk about 24

controls at this point or do we want to start wrapping 25

406


up? 1

MR. DIKEMAN: You're mailing this out, 2

right? 3

MS. ARMSTRONG: We are going to mail it out. 4

MR. DIKEMAN: We can read it. 5

MS. ARMSTRONG: That's correct. The only 6

benefit is, do you want Sanaee to at least start 7

presenting and get a little bit of feedback for the 8

next 10 minutes or 15 minutes or so. I don't think 9

it's going to take 30 minutes to wrap up. 10

MR. DIKEMAN: Sure. 11

MS. IYAMA: I'll try to do this in like five 12

minutes. 13

So we also looked at, and that's not in the 14

NODA because it wasn't used in there, developing 15

default values to model motor and controls, so the 16

whole motor control, motor and VFD part. And it looks 17

like this, and it's similar to what was done and used 18

in the pumps rulemaking. So, if you want more details, 19

you can go into that docket, but it's the same concept 20

where the losses of the motor plus VFD, which is the LD 21

here, equals the full load losses of the motor only 22

times a factor. That factor is a polynomial equation 23

with different coefficients depending on the size of 24

the motor. So it's a pretty simplified model. 25

407


Again, I think in the draft 207 there's 1

another way of doing this. I don't think -- 2

MR. DIKEMAN: Steve Dikeman. We've combined 3

motor and VFD into a single aggregate loss in the most 4

recent writeup. 5

MS. IYAMA: Yeah, so it's similar. I think 6

the form of the equation and the way it's calculated, 7

it's a bit different. 8

MR. SMILEY: Question. This is Bill, Trane. 9

This only applies to full load. 10

MALE VOICE: No. 11

MR. SMILEY: It applies to part load too? 12

Where's the part load? 13

MS. IYAMA: So the little i that's like in 14

subscript means the load. So the first thing -- so 15

here, this component here, and I'm going to increase 16

the -- 17

MR. SMILEY: So it's the motor full load 18

times -- 19

MS. IYAMA: It's the motor VFD -- 20

MR. SMILEY: -- a de-rate factor as a 21

function of what? 22

MS. IYAMA: So that's the load of the motor, 23

of the VFD, sorry. Of the motor, sorry. 24


408


MS. IYAMA: So, in other words, the part 1

load losses of your motor plus VFD equal a certain 2

factor times the full load losses of the motor. 3

MR. SMILEY: So you're saying that all the 4

data that this is based on is available somewhere as 5

well? 6

MS. IYAMA: Maybe not the disaggregated 7

data, but there's more details in that docket that's 8

referenced on that slide. 9

MR. SMILEY: Okay, thanks. 10

MR. WHITWELL: So, if you can send that to 11

us and we don't have to go looking for it, that would 12

be nice. 13

MS. IYAMA: Sure. Yeah. I'll dig up the -- 14

and then -- so I think that's all. These are just sort 15

of summary tables of all the things we've discussed. 16

And then on the test procedure -- 17

MR. SMITH: Sanaee, this is Wade Smith. I 18

apologize. Could you just describe once again the 19

source of the default values for -- 20

MS. IYAMA: Motor and drive? 21

MR. SMITH: Yeah, motor and drive, the 22

combination. 23

MS. IYAMA: That was based on a combination 24

of data, some from motors and VFDs that DOE tested at 25

409


different loads, some that one manufacturer provided. 1

MR. SMITH: And it was commented earlier 2

that the efficiency performance of the motor and drive 3

in combination, it depends upon the settings, the setup 4

and the settings, and that they're easy to change. So 5

is there some caveat about these default values in 6

terms of how the drive is set up? 7

MS. IYAMA: Currently, no. 8

MR. SMILEY: Well, that would -- Bill with 9

Trane. That would go along with that, would be the 10

type of inverter and the type of motor would play into 11

that as well, I mean, you know, if you really needed to 12

dig into the minute details. You're coming up with a 13

default relationship. 14

MR. SMITH: And the default, you referred to 15

a docket. Is there a docket where this question is 16

being dealt with? No? 17

MS. ARMSTRONG: So I think the answer is 18

kind of. We can refer you to it. 19

MS. IYAMA: I put the docket number. 20

MR. SMITH: It's in the pump rule? 21

MS. ARMSTRONG: Kind of. So I think that, 22

and I wish some motor guys were here today. So the 23

motor guys are in the process kind of of coming up with 24

this motor drive controls type methodology. It was 25

410


based somewhat on that. It's generally similar to that 1

whole exercise they are going through. 2

Why don't we do this. We will commit to 3

circulating the explanation of that so you guys can 4

look at it, and obviously we welcome your feedback on 5

it here. It's open. So let's do that. We'll just 6

pull it off the docket. It's easier that way. 7

MS. IYAMA: Yeah. There's the AMCA 207 way, 8

there's this way, there's the NEMA way, so we can just 9

compare them all. 10

MS. ARMSTRONG: The NEMA way is really close 11

to what we did in pumps. It wasn't done when we had to 12

go out with pumps, but we had the draft. 13

MR. WAGNER: Are we able to get the AMCA 207 14

draft then? 15

MS. ARMSTRONG: I'm circulating that. 16

MR. HAUER: I think AMCA 207 was part of the 17

NODA response, was it not? 18

MS. ARMSTRONG: Yeah. So the answer is it's 19

in the docket. 20

MS. IYAMA: Okay. So quickly, test 21

procedure, I have three slides really. So we need -- 22

why do we need -- we need a test method to either get 23

to the shaft input power if we're using a calculation-24

based method or we need to be able to measure the 25

411


electrical input power at the arrow number two here on 1

the slide. 2

One approach that we've been considering is 3

to use AMCA 210 as the basis for the DOE test 4

procedure. Question, is there any modifications 5

necessary to ensure that every time someone does the 6

test we get the same answer? These are lists, and we 7

don't need to go through all of them, but these are the 8

sort of -- 9

MS. ARMSTRONG: Yeah. So we are going to 10

leave you with this as homework for next week because 11

what we have done is taken a deep dive into AMCA 210 12

and we had some questions. 13

MS. IYAMA: And some of those are not very 14

detailed. It's just to get the conversation started 15

and sort of give you the sort of topics that we're 16

interested in and get your feedback, and we can get in 17

more details at the next meeting. 18

MR. SMILEY: This is Bill with Trane. Are 19

you asking for additional test methodologies that may 20

already be out there that people use for lots of 21

different types of products to be brought forward as 22

well? We haven't settled on AMCA 210. 23

MS. ARMSTRONG: Sure. I mean, I think at 24

this point you're right, DOE hasn't put a proposal out, 25

412


so at this point it's open-ended. If there are certain 1

categories of fans or equipment classes of fans that 2

you believe should be tested with a different 3

methodology, we're open to having that discussion. One 4

thing that will be important is that at the end of the 5

day ratings are generated in an equitable manner. So 6

we'll need to understand -- and if they're not, why 7

they're not generated in an equitable manner, so if 8

certain equipment classes need to be tested in a 9

different manner. 10

MR. SMILEY: Well, the reason I bring it up 11

is there are existing tests that are already being 12

performed on a lot of the equipment that will probably 13

be covered by this, and nobody wants to do additional 14

testing just for the heck of it. That's why I bring 15

that up. 16

MS. ARMSTRONG: No, but I would say one of 17

the purposes of this regulation obviously isn't to make 18

everyone retest the testing they had already done but 19

make sure testing is done in the same manner. So if 20

there are some differences or nuances it would be great 21

to have a discussion around what those are. Okay? 22

So, with that, I think Sanaee is done. So I 23

do want to go back and do some kind of close-out items. 24

I'm happy to open the floor to others for closure, but 25

413


I also have tried to come up with a list over the past 1

couple of hours of things that at least were in my 2

mind. You guys can remind me if I missed anything. 3

And I think it would be a good idea to put some dates 4

around these. I will say over half of them are for 5

DOE, so obviously we're going to have our work cut out 6

for us over the next couple of days. 7

But one thing. First off, AMCA was going to 8

provide feedback on the compressor cutoff in the 9

definition. When can you do that? 10

MR. SMITH: Before the next meeting or at 11

the next meeting. 12

MS. ARMSTRONG: Great. Thank you. 13

MR. SMITH: Meaning that we agree with it or 14

don't agree with it? 15

MS. ARMSTRONG: So either you agree with it 16

or if you don't agree with it an alternative solution. 17

MR. SMITH: Okay. 18


So next was AHRI's data request. There's 20

two parts to the data request, so, Karim, I'm looking 21

to you. When do you at least think you'll be able to 22

respond to part one, which is just simply a list of 23

different categories of fans, what type of equipment 24

they may or may not go into, in this case will go into, 25

414


and are those fans regulated? So is the energy 1

consumption of those fans that are in embedded 2

equipment regulated in another manner, system metric or 3

otherwise? 4

MR. AMRANE: At the next meeting? 5


And so the next one is DOE's going to send 7

out red lines of all these. We may do this in phases. 8

So we may do tomorrow's first and then this one will 9

follow in a day or two. One of the things we've 10

already done is include the EU comparison of 11

definitions that was asked for, so you're going to see 12

that in the red lines that we sent out for you guys to 13

all mull over, so be on the lookout for that. We'll 14

commit to sending one out today and one out tomorrow. 15

It's going to be a fun night. 16

Okay. So one of the things I think for all 17

of us is to review the definitions for the equipment 18

classes and the testable configurations. So, when we 19

send out those red lines, you're going to see the 20

reflection of what is this testable configuration for a 21

variety of different things. We'll also include the 22

examples, so we'll want to get back with feedback on 23

that if anybody has any or obviously agreement on that 24

at the next meeting. 25

415


So one of the things that I think the 1

embedded product manufacturers, including AHRI, were 2

going to bring to the table, and I'm just asking for a 3

time frame, was some type of counterproposal or pathway 4

forward for some of the embedded products in terms of 5

how they would like the working group to consider them 6

and consider treatment of them. So my ask of you guys 7

is, what is your timing on that one? 8

MS. SHEPHERD: Well, I think we were 9

planning for the next meeting, so I would say within 10

the next two meetings you're going to have or the next 11

week of meetings. 12

MS. ARMSTRONG: Okay. So I think the first 13

part of June is our next meeting? 14

MS. SHEPHERD: Right. 15

MS. ARMSTRONG: Thank you. If you want to 16

add, go ahead. 17

MR. JASINSKI: Yeah. It's not an additional 18

to do, but it's related to both of those. In addition 19

to revised definitions for testable configuration as 20

part of that exercise or as part of any proposal from 21

AHRI, I think it would also be useful to bring up your 22

own examples of fans that can be used or come up with 23

your example fans that test the definitions in the way 24

that you want to test them and as part of justification 25

416


for tweaks or as additional examples to be discussed at 1

the working group. I know some people mentioned that 2

they had fans in mind while we were talking about that 3

separately. So, if you have fans that you're 4

particularly interested in knowing how they would be 5

impacted by the definitions that are being proposed, 6

those examples would be really helpful. 7

MS. ARMSTRONG: Okay. So the next one I 8

have for us is to circulate the pump motor variable 9

speed info in terms of how we got to that. There is a 10

memo in the pump docket, so we'll just pull it out and 11

circulate it so you have that as background. 12

The next thing we'll do is send out a link 13

to the draft AMCA 207 that's also already in this 14

docket, but just so ease of use. 15

The last one -- we can do all these things 16

tonight, but the last one is the feedback on what we 17

just presented in terms of we went through some sticky 18

issues today in terms of an approach for a test 19

procedure. We listed a whole page of questions we have 20

relating to AMCA 210. We have questions relating to 21

nominal values versus tested values, nominal values for 22

specific categories of motors, nominal values versus in 23

control and drive systems. So we'd really like your 24

feedback on certain of those, so you'll see some red 25

417


lines coming out probably tomorrow, so we'd like some 1

feedback on that. So does everyone around the table 2

think they can be prepared to discuss that at the next 3

meeting? Okay, I see nods. 4

And last but not least, we are going to be 5

prepared at the next meeting to present the NODA 6

analysis, so we're going to walk through the 7

spreadsheets. If anyone has an issue with downloading 8

the spreadsheets, using the spreadsheets, come talk to 9

us or send us an email. They've been up there. They 10

should be able to be used, but like I said, if there's 11

an issue, let us know. We'll walk through them at the 12

next meeting so we can talk about analysis. 13

If we make some decisions on this stuff that 14

impacts the analysis, we can then revise after the next 15

meeting, but we want to be in a position to move 16

forward with some of the analysis discussions so 17

everyone will have a good basis for that understanding. 18

okay? 19

MR. WAGNER: Ashley, I noticed some of the 20

pull-down boxes and stuff like that weren't functional. 21

Is that because of the way I downloaded it or -- 22

MS. ARMSTRONG: I don't know. It could be 23

your security settings that are off, like not -- 24

MR. BUBLITZ: I have a screen snapshot. 25

418


MS. ARMSTRONG: You'll show me. Okay, no 1

worries. We'll get it fixed if there's an issue. 2

MR. WAGNER: Okay. Like the direct drive, 3

belt drive, and some of the others, it wouldn't pull 4

down. 5

MS. ARMSTRONG: We'll look at it. No 6

problem. If we need to recirculate a new version, we 7

can. No problem. 8

MR. SMITH: I had the same issue. 9

MS. ARMSTRONG: Okay. Maybe you just have 10

to be on a DOE computer, so you all can come visit to 11

work on these spreadsheets. 12

So that's all I had. Does anybody have 13

anything else? 14

MR. SMILEY: Yes. Bill, Trane. You said 15

you were going to give us the information you have on 16

the drive motor interaction, which was the last thing 17

you showed on the efficiency de-rate for a variable 18

speed. 19

Also, previous to that I think I'd asked for 20

if you had any data on just the motor and the load 21

point on the motor efficiency reduction, because what 22

you showed was a nominal curve to apply to everything, 23

but what I was interested in is if you had the backup 24

data that might show the function of other variations. 25

419


MS. IYAMA: So I would need to check if 1

we're allowed to share the data. I mean, we have the 2

data, but I know for sure that the one for the drive in 3

VFD was confidentially -- 4

MR. SMILEY: Yeah, I figured that probably 5

was. 6

MS. IYAMA: So -- 7

MR. SMILEY: I just wanted to compare it to 8

stuff I have and I use. 9

MS. ARMSTRONG: So anybody want to say 10

anything else? 11

MR. AUTH: Just one more comment about the 12

NODA spreadsheet. Chris Auth, Baltimore Aircoil. 13

Right now it's set up for total efficiency and we did 14

discussions today with static efficiency. Is there 15

going to be a revision to that before next meeting or 16

just use it the way it is? 17

MS. ARMSTRONG: So we have the results for 18

the next meeting for sure. We can have clean 19

spreadsheets by the next meeting and we'll see how 20

early we can have the clean spreadsheets to you guys 21

before the next meeting. But yeah, we have the results 22

at least to show you for the next meeting or to 23

circulate before the next meeting so you can see them. 24

No problem. 25

420


Anything else? 1

MS. WALTNER: Yeah. This is Meg, just 2

really quickly. Earlier some of us were asking about 3

the times of the meetings. I think we confirmed them 4

last time, but I think it would be helpful to just go 5

over again the start and end times we agreed to for the 6

meetings. I don't know if we have that. 7

MS. ARMSTRONG: Yeah. I'll send out the 8

schedule as part of the email for the next one. 9

You can go off the record now. 10

(Whereupon, at 2:48 p.m., the meeting in the 11

above-entitled matter was concluded.) 12

// 13

// 14

// 15

// 16

// 17

// 18

// 19

// 20

// 21

// 22

// 23

// 24

// 25

421


REPORTER'S CERTIFICATE

DOCKET NO.: N/A

CASE TITLE: ASRAC Fans and Blowers

Working Group Meeting

HEARING DATE: May 19, 2015

LOCATION: Washington, D.C.

I hereby certify that the proceedings and

evidence are contained fully and accurately on the

tapes and notes reported by me at the hearing in the

above case before the U.S. Department of Energy, Office

of Energy Efficiency & Renewable Energy.

Date: May 19, 2015

David W. Jones Official Reporter Heritage Reporting Corporation Suite 206 1220 L Street, N.W. Washington, D.C. 20005-4018

Documents

heritage reporting corporation